the reference point effect, m&a misvaluations and merger...
TRANSCRIPT
Durham E-Theses
The Reference Point E�ect, M&A Misvaluations and
Merger Decisions
LI, ZHENLONG
How to cite:
LI, ZHENLONG (2017) The Reference Point E�ect, M&A Misvaluations and Merger Decisions , Durhamtheses, Durham University. Available at Durham E-Theses Online: http://etheses.dur.ac.uk/12225/
Use policy
The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission orcharge, for personal research or study, educational, or not-for-pro�t purposes provided that:
• a full bibliographic reference is made to the original source
• a link is made to the metadata record in Durham E-Theses
• the full-text is not changed in any way
The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
Please consult the full Durham E-Theses policy for further details.
Academic Support O�ce, Durham University, University O�ce, Old Elvet, Durham DH1 3HPe-mail: [email protected] Tel: +44 0191 334 6107
http://etheses.dur.ac.uk
The Reference Point Effect,
M&A Misvaluations and Merger
Decisions
Zhenlong Li
Principal Supervisor: Dr. Jie (Michael) Guo
A Thesis Presented for the Degree of Doctor of
Philosophy in Finance
Durham University Business School
Durham University
Mar 2017
I
The Reference Point Effect,
M&A Misvaluations and Merger Decisions
Abstract
This thesis investigates the reference point effect on M&As. Prospect theory proposes that the
reference dependence bias is firmly rooted in people’s minds, affirming that people rely heavily
on a piece of relevant information while making decisions. This thesis applies this reference
point to M&As, showing that M&A participants are subject to reference-dependence bias. The
reference point theory explains the M&A motives in a new way. The results of this thesis offer
important implications to M&A practitioners.
Baker et al. (2012) suggest that the target’s reference price enhances the target’s bargaining
power in the U.S. market, inspiring the author’s thinking that the reference point effect is likely
to be reinforced in a competitive market and in the scenario where the bidder is in the face of
considerable information barriers. In this pursuit, Chapter 3 studies the reference point effect
on a sample of public acquisitions involving a U.K. target. Using the target 52-week high as
the reference price, a positive relationship emerges between the reference price and the offer
premium. Further, there is evidence of overpayment among domestic bidders whereas little
evidence among cross-border bidders, indicating that the market acquiesces the payment
according to the target reference point among cross-border bidders. Evidence that reference-
dependence bias serves as the bargaining power is confirmed outside the United States.
However, the real M&A motive of the bidder is unlikely to be revealed solely depending on
the target reference point. Therefore, Chapter 4 extends Baker et al.’s paper (2012) by adding
both the target and the bidder reference points. A Relative Reference Point (RRP) is proposed
for M&A misvaluations as per Shleifer and Vishny (2003), indicating the extent to which the
bidder is more overvalued relatively to the target. The joint reference point effects can explain
the motivation of why bidders paying high offer premiums. The results obtained in this chapter
show that bidders are likely to pay with stocks when RRP increases, suggesting that bidders
accelerating the process of overvaluation dilution is a sign of bidder overvaluation. In addition,
the offer premium increases with the RRP, indicating the motive of diluting overvaluation is to
be seen by the target, leading targets to bargain over high offer premiums. When assessing the
long-term performance of stock bidders, it became evident that bidders time the market by
paying stocks according to the RRP.
A direct implication of prospect theory is that people seek risk in the face of loss gauged by a
reference point. Chapter 5 proposes the bidder reference point to assess the market anticipation
effect. Investors suffer mental loss when their firm’s current performance is significantly below
its best recently achieved performance. They anticipate that the firm will take risks to turn the
table round. It was evident that the firm is rewarded with positive market reactions when it
takes risks according to the market’s anticipation. In a further analysis, it also emerged that
managers taking the market-anticipated risks exhibit greater efforts in the processes of M&A
negotiation and post-merger integration, reflected in the low offer premiums paid to the target
shareholders and positive long-term abnormal returns earned.
This thesis has empirically investigated many implications of the reference point effect in the
M&A context. The main aim of this thesis is to simplify M&A decisions for managers in order
to structure M&A effectively.
II
No part of this thesis has been previously submitted for a degree in this and other university.
The copyright of this thesis rests with the author or the University to which it was submitted. No quotation from
it, or information derived from it should be published without the written consent of the author or the University,
and information derived from it should be acknowledged.
III
To my parents
IV
Acknowledgements
There are many people have helped me considerably in the completion of my Ph.D studies.
First, I would like to give the greatest gratitude and appreciation to my principal supervisor Dr.
Jie (Michael) Guo who brought me into academia. He offered me valuable advice and
knowledge throughout my Ph.D journey. He also encouraged me to attend high quality
academic conferences and recommended me to work in the financial institution. Those
experiences are of great benefits to my professional career and research development. My Ph.D
journey would not go thus far if without his encouragement and generous support. I would like
to extend my thanks to Michael. It is my greatest fortune to be part of his Ph.D team.
I am also grateful to Dr. Panagiotis Andrikopoulos, the co-author for one of my working papers,
for his advice in improving the quality of my thesis, and I would also like to express my
gratitude to Dr. Evangelos Vagenas-Nanos who offered me many thoughts to improve the
fourth Chapter of this thesis.
A huge thank you to my deeply beloved parents, my father Li, Yong and my mother Zheng,
Defeng, whose love to me are unconditional. They funded my Ph.D studies and encouraged
me to move on with a brave and humble heart. There are no words in this world could describe
how much I love you both.
Special thanks to those who have helped me in this wonderful journey. Dr. Kai Lisa Lo, Dr
Yichen Li and Xiaofei Xing, who offered me great support in research. Rev. Lawrence Lo,
whose words of wisdom taught me to deal with the difficulties of life. And Dr. Eason Chan,
whose music inspired me. A number of people, whose names though not listed in the
acknowledgements, their cares to me are unforgettable and the motivation towards the
completion of my Ph.D studies.
Zhenlong Li
14th Feb, 2017
V
Table of Contents
ABSTRACT ............................................................................................................................... I
ACKNOWLEDGEMENTS ..................................................................................................... IV
CHAPTER 1: INTRODUCTION .............................................................................................. 1
CHAPTER 2: LITERATURE REVIEW ................................................................................. 16
2.1. M&A motives ................................................................................................................... 17
2.1.1. Synergies ........................................................................................................................ 18
2.1.2. Agency problems ............................................................................................................ 22
2.1.3. Behavioural finance theories ......................................................................................... 25
2.1.3.1. Managerial overconfidence ........................................................................................ 26
2.1.3.2. Investors’ overreactions and underreactions ............................................................. 30
2.1.3.3. The misvaluation hypothesis ....................................................................................... 32
2.2. M&A process .................................................................................................................... 38
2.2.1. Selecting a target ........................................................................................................... 38
2.2.2. Paying for a target ......................................................................................................... 41
2.2.2.1. M&A offer premiums .................................................................................................. 42
2.2.2.2. Method of payment hypotheses ................................................................................... 45
2.2.2.2.1. Tax considerations ................................................................................................... 46
2.2.2.2.2. Signalling hypothesis ............................................................................................... 47
2.2.2.2.3. Capital structure and corporate control .................................................................. 49
2.2.2.2.4. Behavioural finance ................................................................................................. 51
2.3. M&A performance ............................................................................................................ 55
2.3.1. Short-term performance ................................................................................................. 56
2.3.2. Long-term performance ................................................................................................. 60
2.3.2.1. Stock performance ...................................................................................................... 60
2.3.2.2. Operating performance ............................................................................................... 62
2.3.3. Factors influencing M&A performance ......................................................................... 65
2.3.3.1. The method of payment ............................................................................................... 65
2.3.3.2. Target status................................................................................................................ 68
2.3.3.3. Merger relatedness ..................................................................................................... 69
2.3.3.4. Merger attitudes .......................................................................................................... 70
2.3.3.5. Merger choice ............................................................................................................. 70
VI
2.3.3.6. Size and relative size ................................................................................................... 72
2.3.3.7. Market-to-book ratio ................................................................................................... 74
2.4. Prospect theory.................................................................................................................. 76
2.4.1. Diminishing sensitivity ................................................................................................... 77
2.4.2. Reference dependence .................................................................................................... 77
2.4.3. Loss-aversion tendency .................................................................................................. 78
2.4.4. Prospect theory and M&As ............................................................................................ 79
CHAPTER 3: REFERENCE POINT THEORY ON CROSS-BORDER AND DOMESTIC
M&A DEALS: U.K. EVIDENCE ........................................................................................... 81
ABSTRACT ............................................................................................................................. 82
3.1. INTRODUCTION ............................................................................................................ 83
3.2. LITERATURE REVIEW ................................................................................................. 90
3.2.1. Cross-border M&As in the U.K. .................................................................................... 90
3.2.2. The reference point effect on the U.K. ........................................................................... 93
3.2.3. The target 52-week high as a reference point price ...................................................... 96
3.2.4. M&A offer premiums ..................................................................................................... 98
3.2.5. Reference point price and the offer premium .............................................................. 103
3.3. HYPOTHESIS DEVELOPMENT.................................................................................. 105
3.4. DATA AND METHODOLOGY .................................................................................... 108
3.4.1. Data.............................................................................................................................. 108
3.4.2. Methodology ................................................................................................................ 114
3.4.2.1. Piecewise linear regression and local polynomial smooth procedure ..................... 114
3.4.2.2. Short-term event study .............................................................................................. 117
3.4.2.2.1. Market-adjusted model .......................................................................................... 117
3.4.2.2.2. Market Model ......................................................................................................... 119
3.4.2.3. Multivariate analysis ................................................................................................ 120
3.4.2.4. Two Stage Least Squares (2SLS) .............................................................................. 121
3.5. EMPIRICAL RESULTS ................................................................................................. 122
3.5.1. The reference point effect on the offer premium .......................................................... 122
3.5.2. The reference point effect on bidder announcement returns ....................................... 130
3.6. ROBUSTNESS CHECKS .............................................................................................. 132
3.6.1. The target 52-week high as a reference point .............................................................. 132
3.6.2. The reference point effect on the offer premium by market conditions ....................... 133
3.6.3. The reference point effect on the offer premium across different subsamples ............ 134
VII
3.6.4. The reference point effect on bidder announcement returns ....................................... 135
3.7. CONCLUSION ............................................................................................................... 136
CHAPTER 4: RELATIVE REFERENCE PRICES AND M&A MISVALUATIONS ........ 159
ABSTRACT ........................................................................................................................... 160
4.1. INTRODUCTION .......................................................................................................... 161
4.2. LITERATURE REVIEW ................................................................................................ 171
4.2.1. The misvaluation hypothesis ........................................................................................ 171
4.2.2. Reference point theory of M&As .................................................................................. 181
4.3. HYPOTHESIS DEVELOPMENT.................................................................................. 186
4.4. DATA AND METHODOLOGY ..................................................................................... 188
4.4.1. Data.............................................................................................................................. 188
4.4.2. Summary statistics ....................................................................................................... 192
4.4.3. Methodology ................................................................................................................ 193
4.4.3.1. Relative reference point (RRP) and offer premiums ................................................. 193
4.4.3.2. Logistic regressions .................................................................................................. 195
4.4.3.3. Classification of high-, neutral- and low-valuation markets .................................... 196
4.4.3.4. Short-term method .................................................................................................... 197
4.4.3.5. Long-term method ..................................................................................................... 197
4.4.3.5.1. Market-adjusted model .......................................................................................... 197
4.4.3.5.2. Size-adjusted model ............................................................................................... 198
4.4.3.5.3. Bootstrapped t-statistics......................................................................................... 199
4.4.3.6. Multivariate analysis ................................................................................................ 200
4.5. EMPIRICAL RESULTS ................................................................................................. 200
4.5.1. The RRP effect on the probability of using stocks ........................................................ 201
4.5.2. The RRP effect on the offer premium ........................................................................... 203
4.5.3. Do bidders who focus on RRP overpay for the target? ............................................... 207
4.5.4. Do all stock-financed acquisitions driven by RRP protect the wealth of long-term
shareholders? ......................................................................................................................... 208
4.6. ROBUSTNESS CHECKS .............................................................................................. 210
4.6.1. Endogeneity issues ....................................................................................................... 210
4.6.2. The effect of the RRP on the probability of using stocks.............................................. 211
4.6.3. The effect of the RRP on the offer premium ................................................................. 212
4.6.4. Do bidders who focus on RRP overpay for the target? Testing with the market-adjusted
model ...................................................................................................................................... 213
VIII
4.7. CONCLUSION ............................................................................................................... 213
CHAPTER 5: MARKET ANTICIPATION AND MERGER DECISIONS ......................... 236
ABSTRACT ........................................................................................................................... 237
5.1. INTRODUCTION .......................................................................................................... 238
5.2. LITERATURE REVIEW ............................................................................................... 245
5.2.1. Reference point theory and gambling .......................................................................... 245
5.2.2. Reference point theory and risk-taking ........................................................................ 248
5.2.3. Risk-taking and M&A outcomes .................................................................................. 253
5.3. HYPOTHESES DEVELOPMENT ................................................................................ 258
5.4. DATA AND METHODOLOGY .................................................................................... 260
5.4.1. Data.............................................................................................................................. 260
5.4.2. Methodology ................................................................................................................ 263
5.4.2.1. Target risks and the bidder 52-week high................................................................. 263
5.4.2.2. Short-term method .................................................................................................... 265
5.4.2.3. Long-term method ..................................................................................................... 265
5.4.2.4. Multivariate analysis ................................................................................................ 266
5.5. EMPIRICAL RESULTS ................................................................................................. 267
5.5.1. Does market anticipation affect M&A decisions? ....................................................... 267
5.5.2. Do market-anticipated risks create value? .................................................................. 268
5.5.3. What is the source of bidder performance improvement? ........................................... 269
5.5.4. Do market-anticipated risks push the firm to work hard? ........................................... 270
5.6. ROBUSTNESS CHECKS .............................................................................................. 271
5.6.1. Various reference point candidates based on firm level. ............................................. 271
5.6.2. Subsample tests ............................................................................................................ 273
5.6.3. Alternative proxies for target risks .............................................................................. 273
5.6.4. Risk-taking and M&A performance ............................................................................. 274
5.7. CONCLUSION ............................................................................................................... 274
CHAPTER 6: CONCLUSIONS ............................................................................................ 292
6.1. MAIN FINDINGS .......................................................................................................... 293
6.2. IMPLICATIONS, LIMITATIONS AND RECOMMENDATIONS FOR FUTURE
RESEARCH ........................................................................................................................... 298
REFERENCES ...................................................................................................................... 301
IX
List of Tables
Table 3.1: Summary statistics for M&A sample .................................................................... 138
Table 3.2: Distribution of bidder nations .............................................................................. 139
Table 3.3: Summary statistics for variables .......................................................................... 140
Table 3.4: Summary statistics for variables by cross-border and domestic subsamples ...... 141
Table 3.5: The piecewise linear regressions of offer premiums on the target 52-week high . 142
Table 3.6:The piecewise linear regression of offer premiums on the target 52-week high of
cross-border and domestic acquisitions into the United Kingdom ........................................ 143
Table 3.7: The OLS regression of bidder announcement returns on offer premiums ............ 144
Table 3.8: Robustness test: Controlling for other relevant price measures ........................... 145
Table 3.9: Robustness test: Controlling for target past returns ............................................. 146
Table 3.10: Robustness test: Reference point effect on offer premiums by market-wide
valuation ................................................................................................................................ 147
Table 3.11: Robustness test: U.K. and U.S. bidder subsamples............................................. 148
Table 3.12: Robustness test: Reference point effect on offer premiums by the method of
payment .................................................................................................................................. 149
Table 3.13: Robustness test: Reference point effect on offer premiums for diversified and
undiversified M&A deals ....................................................................................................... 150
Table 3.14: Robustness test: Reference point effect on offer premiums for tender offers and
mergers ................................................................................................................................... 151
Table 3.15: Robustness test: Market model for 5-day CARs ................................................. 152
Table 3.16: Robustness test: Market-adjusted model for 3-day CARs .................................. 153
Table 3.17: Robustness test: Market model for 3-day CARs ................................................. 154
Table 4.1: Summary statistics for M&A sample..................................................................... 215
Table 4.2: Summary statistics for variables ........................................................................... 216
Table 4.3: The effect of RRP on the probability of using stocks ............................................ 217
Table 4.4: The effect of RRP on the probability of using stocks under different market-wide
valuations ............................................................................................................................... 218
Table 4.5: The effect of RRP on offer premiums .................................................................... 220
Table 4.6: The effect of RRP on offer premiums over time .................................................... 221
Table 4.7: The effect of RRP on offer premiums by valuation periods .................................. 222
X
Table 4.8: Univariate analysis by different RRP groups ........................................................ 223
Table 4.9: OLS regressions of offer premiums on RRP: By RRP subsamples ....................... 224
Table 4.10: Do bidder focusing on RRP overpay for the target? .......................................... 225
Table 4.11: Do all stock-financed acquisitions driven by RRP protect the wealth of long-term
shareholders? ......................................................................................................................... 226
Table 4.12: Endogeneity issues .............................................................................................. 227
Table 4.13: Variables correlation matrix ............................................................................... 228
Table 4.14: Robustness check: RRP effect on the probability of using stocks ...................... 229
Table 4.15: Robustness test: The effect of RRP on the probability of using stocks: Between
payment methods .................................................................................................................... 230
Table 4.16: Robustness test: The effect of RRP on offer premiums: By payment methods ... 231
Table 4.17: Robustness test: The effect of RRP on offer premiums: By deal types ................ 232
Table 4.18: Robustness test: The effect of RRP on offer premiums: By deal choice ............. 233
Table 4.19: Robustness test: The effect of RRP on offer premiums: By deal completion ...... 234
Table 4.20: Robustness test: market-adjusted model ............................................................. 235
Table 5.1: Summary statistics for M&A sample..................................................................... 276
Table 5.2: Summary statistics for variables ........................................................................... 277
Table 5.3: Risky and less risky acquisition sample ................................................................ 279
Table 5.4: Reference point effect on M&A decisions ............................................................ 280
Table 5.5: OLS regressions of bidder CARs on market-anticipated risks ............................. 281
Table 5.6: Whether market-anticipated risks increase offer premiums? ............................... 282
Table 5.7: Do market-anticipated risks push the firm to work hard? .................................... 283
Table 5.8: Robustness tests: Various reference candidates in relation to the firm level ........ 284
Table 5.9: Robustness test: Do the market returns or peer pressures play a role in the firm’s
risk-taking? ............................................................................................................................ 286
Table 5.10: Robustness test: Subsample tests ........................................................................ 288
Table 5.11: Robustness test: OLS regressions of target idiosyncratic risk on the bidder
reference point ....................................................................................................................... 289
Table 5.12: Robustness test: OLS regressions of bidder CARs on market-anticipated risks . 290
Table 5.13: Robustness test: OLS regressions of BHARs on market-anticipated risks ......... 291
XI
List of Figures
Figure 3.1: Distribution of number of deals by bidders ........................................................ 155
Figure 3.2: Distribution of value of deals by bidder types .................................................... 156
Figure 3.3: Offer premiums relative to the target 52-week high ........................................... 157
Figure 3.4: Nonlinear relation between offer premiums and the target 52-week high .......... 158
XII
“We're blind to our blindness. We have very little idea of how little we know. We're not
designed to know how little we know.”
Daniel Kahneman
1
CHAPTER 1: INTRODUCTION
Chapter 1. Introduction
2
The central purpose of this thesis is to examine the reference point effect on the causes and
consequences of Mergers and Acquisitions (M&As). In Collins’ English Dictionary: the
reference point is ‘a fact forming the basis of evaluation or assessment’. The notion of the
reference point was first noted in the work of Tversky and Kahneman (1974), and developed
as an important implication of Kahneman and Tversky’s prospect theory (1979). Prospect
theory is a descriptive theory depicting how individuals make decisions under risks. A series
of implications were built in relation to the reference point, a specific point that gauges people’s
mental feelings about gains and losses. Prospect theory suggests that people have a strong loss-
aversion tendency relative to a reference point, suggesting that people are unwilling to suffer
losses and thus seeking many risks when experiencing loss, while, on the other hand, they avoid
taking too many risks when enjoying gain. Feelings about the loss are more intense than those
relating to gain, and feelings tend to be stronger nearer the reference point than further away
from it in both the loss and the gain domains.
M&As serve as an ideal testing ground for prospect theory for four proposed reasons. First,
M&As are major corporate investment activities associated with large risks. Prospect theory
addresses how managers make decisions in consideration of uncertainty, which is a suitable
theory to explain the M&A motive. M&As decisions are complex as they normally involve a
negotiation process that requires superior skills and considerable efforts by bidder managers
who are committed to completing the M&A deal to create the wealth for shareholders. It
remains largely unexplored how much a bidder manager should pay for the target. Valuing
targets is a very subjective task since it rests upon a large number of assumptions that make the
decision-making process hard to be observed by outsiders. In such a case, the reference point
price, a relevant piece of information, is likely to be employed to assess M&A motive.
Chapter 1. Introduction
3
Secondly, whilst existent behaviour finance theories have explained M&A motives on a basis
of deal synergies, the hubris management hypothesis (Roll, 1986; Hayward and Hambrick,
1997), managerial overconfidence (Malmendier and Tate 2005, 2008), self-attribution bias
(Doukas and Petmezas, 2007) and the misvaluation hypothesis (Shleifer and Vishny, 2003;
Dong et al., 2006), prospect theory explains the M&A motive from a perspective of managerial
decision-making process. This theory explains decision-making from a human’s nature
response, without making any assumptions regarding whether the market or the managers are
rational during the course of the transaction. Given that the outcome of major investment
decisions significantly affects the interests of stakeholders, the reference point price, as it is
easily obtained, is likely to be considered by the managers. The reference point theory of
M&As thus allows the researcher to observe how the reference-dependence bias affects
managerial major corporate investment decisions, such as initiating an M&A deal or valuing
the target.
Thirdly, whilst traditional theories explain the M&A motive from the perspective of the bidder,
reference point thinking could help the researcher to explain M&A motives from the
perspectives of the two firms involved in an M&A deal (i.e. the two firms involved). Because
the agreement of an M&A deal cannot be reached without the consent of the two firms involved,
bidders should consider the willingness for deal acceptance on the target side while making
M&A decisions. The reference point used in this thesis is an item of price information that
reflects the market’s perspective on a firm’s prospects. Managers who make major decisions
should also be concerned about the market’s reactions. Therefore, focusing on the reference
point effect enables a big picture on how investment strategy is formulated and how it is
structured from the perspectives of major participants in M&As. This thesis examines role of
the reference point effect played on various central aspects of M&As, including offer premiums,
Chapter 1. Introduction
4
the method of payment and managerial risk-taking attitude, which significantly affect M&A
outcomes.
Finally, M&As offer the researcher many considerations on how the reference point effect is
reinforced. One possible explanation for people looking at a salient piece of information is
because there exists considerable uncertainty about the information, making the true situation
less likely to be accurately observed. This raises the concern as to whether the reference point
effect is significant in the setting where the level of information asymmetry is high according
to the information asymmetry hypothesis. Addressing this concern, the reference point theory
of M&As was made use of in a sample consisting of both cross-border and domestic bidders’
acquisitions of a U.K. target. It was expected that the reference point effect would be stronger
for cross-border bidders than domestic bidders based on the rationale that domestic bidders,
with sophisticated networking, are more advantageous in gathering information than cross-
border bidders. A common intuition suggests that managers are skilful and have private
information about their firms, and have better understanding than the market in terms of the
deal. Thus, it is expected that managers (i.e. more experienced investors) should be less likely
to be influenced by a single number, such as a reference point price, instead, they tend to take
other relevant factors into consideration, whilst the market (i.e. less experienced investors)
tends to put greater emphasis on the reference point due to limited information and limited time
in which to process the deal. If this is the case, the reference point effect should have little
impact on managers’ decision-making process. However, it may not be the case, as managers
who work for the best interests of the shareholders should provide rationale to shareholders as
to why they initiate an M&A deal and how it is valued, implying that the reference point is
used as an important decision-making tool for shareholders to judge M&A performance. This
Chapter 1. Introduction
5
thesis integrating prospect theory into the M&A testing ground gives the researcher better
knowledge on how different M&A stakeholders understand M&A motives.
Neoclassical M&A theory suggests that managers are rational and thus add the value to the
firm through M&As (Jensen and Ruback, 1983). According to Jensen (1988), M&As are a
process of allocating and integrating resources of a firm so as to maximise the firm’s value.
His view is well supported by Berkovitch and Narayanan (1993) and Hodgkinson and
Partington (2008) who found synergies gains an essential motive for M&As, dominating over
other proposed M&A motives, such as hubris and agency problems. Despite this, it is a well-
documented fact that bidders do not gain in either the short or the long run, whilst targets gain
in M&As (Jensen and Ruback, 1983). The fundamental reason for bidder’s underperformance
is overpayment, which is the offer price paid to the target shareholders largely is greater than
the expected synergies gains following a merger. Bidder shareholders may find the firm has
difficulty generating synergies if high offer premiums are offered to the target shareholders,
since they are directly observed by the market, leading to negative market reactions once the
bid is announced.
One thread in relevant literature contributes the concept of overpayment to the agency conflicts
of interests between managers and shareholders, which arise when managers seize the
opportunity to acquire firms to pursue their own benefits at the expenses of their shareholders.
As a result, bidder managers tend to pay excessively for targets. Jensen (1986) studied the
performance of a bidder with large free cash flow and found that managers tend to engage in
ill-conceived investment projects. In a similar vein, Morck et al. (1990) find managers fail to
maximise the value of the firm, as they tend to engage in value-destroying deals for the
purposes of risk reduction and job security. Fu et al. (2013) found that severe agency problems
Chapter 1. Introduction
6
lead to the firm’s underperformance in the long term. All of these studies indicate that
managerial overpayment destroys the value of a firm.
Another thread in relevant literature concerns overpayment to behavioural finance explanations.
Roll (1986) was the first to develop the hubris management hypothesis that managers tend to
overestimate the synergies gains to be generated by M&As and accordingly overpay for the
target. In this line of thinking, Hayward and Hambrick (1997) found the self-importance of top
managers leads to overpayment. Malmendier and Tate (2008) found overconfidence in
managers having a strong preference for their firm’s stocks tends to lead to the completion of
more acquisitions associated with negative market reactions. Rau and Vermaelen (1998)
contribute such managerial overconfidence to market-to-book value (MTBV), showing that
high MTBV firms tend to be associated with strong stock price momentum due to better
historical performance, thus being overvalued. The positive market reaction therefore
reinforces the managerial illusion of control1, leading to excessive unnecessary acquisitions.
Moeller et al. (2004) ascribe managerial confidence to firm size, suggesting that larger firms
are associated with worse performance than that of smaller firms, reasoning that managers of
larger firms tend to have fewer restrictions on a firm’s resources and such power of autonomy
arguments managerial overconfidence.
Prospect theory offers many additional considerations for overpayment distinct from those in
behaviour finance theories. Given limited time and limited information in which to process the
deal, bidders and targets tend to employ a relevant piece of price information for decision-
making. Following studies that link corporate investment decisions and the firm’s reference
point price (Heath et al., 1999, George and Hwang, 2004, Huddart et al., 2009, Burghof and
1 The tendency that people overestimate their ability to control events.
Chapter 1. Introduction
7
Prothmann, 2011, Baker et al., 2012), the firm’s 52-week highs were employed as reference
points in this thesis, as they are frequently reported by financial media and more likely to create
a deep impression in the investors’ minds. Targets may presumably believe that a price relative
to their 52-week high is a fair price as it reflects their best performance. According to prospect
theory, people have a strong loss-aversion tendency. If losing the control of the firm is
perceived as a mental loss for the target shareholders, an offer price that is closer to the
reference point is more likely to compensate the feelings of such a loss, making the deal more
acceptable. On the other hand, targets encouraged to use its 52-week high to negotiate a decent
offer price with the bidder, as target shareholders believe the firm may well create value without
it being acquired. Edward and Walkling (1985) regard offer premiums as the outcome of the
bargaining process between the bidder and target managements, suggesting that bidders in a
relatively weak position tend to pay high offer premiums, leading lower abnormal returns,
while targets in a relative strong position can demand high offer premiums being translated
into larger positive abnormal returns. Therefore, the use of the reference point reinforces the
target’s bargaining power in exchange for high M&A offer premiums.
Chapter 3 of this thesis examines the role of reference point effect in the U.K. market. The idea
of Baker et al. (2012), who examined the role of the target 52-week high effect on the U.S.
market, was put forward to the U.K. market. Unlike Baker et al. (2012), a sample containing
both the domestic and cross-border bidders were taken into consideration in the United
Kingdom. The reference point effect was expected to have a substantial impact on cross-border
M&As into the U.K. for three proposed reasons. First, U.K. targets may find it easier to use
their 52-week high to negotiate with the bidder given that targets are in the most competitive
market where their bargaining power is more easily to be reinforced and thus being translated
into higher offer premiums (Moeller and Schlingemann, 2005; Alexandridis et al., 2010).
Chapter 1. Introduction
8
Second, U.K. firms have a relatively strong corporate governance system where target
managers are expected to be more likely to serve the best interests of their shareholders. In that
case, target managers tend to use their 52-week high to seek higher offer premiums, which
reconcile the view of their shareholders. Finally, it is expected a stronger tendency for cross-
border bidders than domestic bidders in terms of looking at the target 52-week high as a
reference point given that there is greater information uncertainty regarding cross-border
bidders over domestic bidders.
Using a sample of 451 domestic acquisitions and 155 cross-border acquisitions into the U.K.
between 1985 and 2014, a positive relationship between the target 52-week high and the offer
premium was emerged. The reference point distinguishes the takeover motive of domestic
bidders from that of cross-border bidders. It became evident that offer premiums driven by the
target 52-week high lead to a negative market reaction for domestic bidders, suggesting
evidence of overpayment. However, there is little evidence of overpayment for cross-border
bidders paying according to the target 52-week high. The findings in this research are consistent
with prospect theory.
Chapter 3 makes two distinct contributions to the M&A literature. First, this is the first
examination of the role of the reference point in the context of cross-border M&As. This allows
the researcher to investigate how bidders value targets under information uncertainty. Further,
the reference point theory of M&As distinguishes the motive of domestic bidders from cross-
border bidders acquiring U.K. public targets. Offer premiums paid based on the target reference
point price have a negative impact on bidder announcement returns for the domestic M&A
subsample while there is an insignificant impact on bidder announcement returns for the cross-
border M&A subsample, indicating that shareholders tend to believe that managers of domestic
Chapter 1. Introduction
9
bidders pay too much for the target based on the target 52-week high, while take those of the
cross-border who pay according to the target 52-week high for granted.
Chapter 3 applies the idea of the reference point effect to the U.K. where targets have a strong
bargaining position with an expectation that they are likely to employ the 52-week high as the
reference point in the negotiation table. This chapter finds that the reference point reinforces
the target’s bargaining power in M&As, and important factors that enhance the reference point
effect on M&As. However, it leaves a great deal to think about the role of the bidder reference
point in the context of M&As in consideration of the bidder 52-week high is also an available
item of information for the market to assess the value of the firm around the takeover
announcement date. Whilst Chapter 3 finds that the target 52-week high is a reference point for
how much the bidder should pay for the target, Chapter 4 studies the bidder 52-week high effect,
which gives market information on how much the bidder could pay for the target. Bidder with
a nearness 52-week high tend to create an image of strong profit-generating ability and richness
in financial resources, whilst those whose current price deviates considerably from the 52-week
high tend to experience many difficulties to finance a deal.
In addition, Baker et al. (2012) reason that bidders who pay according to the target 52-week
high are confident in realising synergies above the same level of the target. However, these
authors found that those bidders do not generate synergies. If they can time the market by
selecting quality deals for the purpose of synergies integration, this leaves the researcher of this
thesis to believe that focusing on the target 52-week high alone may not fully account for the
reference point effect on M&As. Therefore, both the target and the bidder 52-week highs were
taken into considerations and examined in Chapter 4, where a new proxy was developed by
including both the target and the bidder reference points. If the market price reflects the view
Chapter 1. Introduction
10
of the market investors to the firm’s valuation, it should be suggested that a firm with the
nearness 52-week high is likely to be overvalued based on the rationale that investors should
be reluctant to bid up the price, fearing that the firm is already overvalued. In contrast, a firm
whose current price deviates a long way from its 52-week high leads the market to believe that
the firm is experiencing bad news and is thus undervalued. Therefore, prospect theory can
accommodate many important implications of the misvaluation hypothesis.
Whilst behaviour finance theories explain value-destroying deals as bidders have a better than
average illusion and overpay the target shareholders accordingly, Shleifer and Vishny (2003)
proposed the misvaluation hypothesis that bidders overpay for the targets with the purpose of
diluting overvalued stocks. This hypothesis rests upon the assumption that the market is
inefficient whereas managers are rational, thus managers are able to exploit mispricing of the
market. However, this motive is not observed by investors with short horizons, resulting in
negative market reactions to the bid announcement. The theoretical model of the misvaluation
hypothesis yields three important implications: first, overvalued firms tend to pay high offer
premiums to the shareholders of a less overvalued firm. Second, overvalued firms prefer stocks
as a means of finance for acquisitions as it accelerates the process of overvaluation dilution.
Finally, offer premiums hedge against the risks of holding stocks in an overvalued market.
Studies reveal that the rationale behind overpayment is that bidders are threatened by the
danger of overvaluation, as addressed in the Jensen’s work (2005).
Shleifer and Vishny’s study (2003) fail to provide any rationale as to why targets would accept
overvalued stocks given that they are as rational as bidders and their firm is not
underperforming prior to an M&A deal. Rhodes-Kropf and Viswanathan (2004) complemented
their view by arguing that both bidders and targets are rational and the reason for making
Chapter 1. Introduction
11
mistakes by accepting too many overvalued stocks is that targets do not have access to private
information about the bidder, resulting in large valuation errors due to an overestimation of
synergies. Rhodes-Kropf et al. (2005), further relaxing the irrational target assumption,
suggesting that targets will also make mistakes in valuing the deal when market-wide valuation
is high. Dong et al. (2006), who employed the P/B and the P/V of both the bidder and target to
examine what drives M&As, also provided supportive evidence that estimation biases of
managers to the firm’s valuation leads to M&As. They note that it is not the true market
misvaluation of the firm but the view of managerial misvaluation drives M&As.
However, Di Guili (2013) challenges this view by proposing the measures for misvaluation.
He found that a high MTBV could lead to predictions that the firm has better investment
opportunity or the firm is significantly overvalued, leading the researcher to believe previous
studies borrowed such valuation measure cannot fully account for the misvaluation hypothesis.
So as to overcome the bias caused by the firm’s fundamental characteristics, Chapter 4 of this
thesis measured the misvaluation of the firms involved by constructing a new proxy, the
relative reference point (RRP), which is the difference between the target and the bidder’s
reference points. The measure makes no assumption as to the true valuation of the firms but
aims to explain the valuation errors driven by the market affecting M&As. It could be advisable
that market should hold a similar view of the valuation of the two firms involved prior to the
announcement date. The measure covers considerations of both the target and bidder sides.
Since the market price is a reflection of the investors’ view of the firm performance, the RRP
thus directly reflects the market’s perception of a firm’s misvaluation. The measure built upon
the reference point captures market reactions and eliminates estimation biases caused by a
firm’s fundamental characteristics.
Chapter 1. Introduction
12
Using a sample of 1,878 U.S. domestic public M&As between 1985 and 2014, it became
apparent that the propensity for using stocks increases when the RRP increases, suggesting that
the relatively more overvalued bidders tend to use stocks as a means of finance for acquisitions.
In addition, the offer premium increases with the RRP, indicating relatively more overvalued
bidders tend to pay higher offer premiums to the less overvalued targets. The RRP takes on a
role of overpayment in the short term, reflected in negative market reactions. However, the
long-term analysis indicates that offer premiums that are associated with acquisitions of higher
RRP tend to generate less negative long-term returns than acquisitions of a lower RRP,
suggesting that managers of more overvalued firms are able to time the market by looking at
the RRP.
Chapter 4 provides three distinct contributions to the M&A literature. First, a new measure for
misvaluation is proposed, based on market perception. The measure establishes a bilateral
valuation framework where both the bidder and the target’s misvaluation are considered. The
measure also eliminates the biases caused by a firm’s fundamental characteristics or any
retrospective and forward-looking information since it reflects the latest market reaction.
Secondly, we provide evidence that both experienced and less experienced investors tend to be
affected by the reference point effect. However, they may interpret the reference point in
different ways. An investor with a short-term horizon tends to contribute the offer premium
paid to the target shareholders with the result of overpayment, whilst a bidder who pay offer
premiums according to the RRP times the market, reflected in the less negative market. Finally,
it is suggested that the RRP is an important indicator of the method of payment. Overvalued
bidders who use stocks rather than cash as a means of finance for acquisitions tend to pay lower
offer premiums. It was established that lower offer premiums are associated with cash instead
Chapter 1. Introduction
13
of stock payments for relatively more undervalued targets. These results offer a new
explanation for the method of payment hypothesis.
One important implication of the reference point is that people tend to seek risks when they
experience loss relative to the reference point and adopt a risk aversion when they enjoy gains
relative to this specific point. Chapter 5 of this thesis directly assessed this implication by using
the bidder 52-week high as the reference point. Since the firm’s 52-week high reflects the
market reaction to the bidder’s performance, the loss relative to a reference point price serves
as a relevant piece of information that signals to the market that risky investment opportunities
should be available. Managers who fail to utilise this will be blamed for not meeting the
commitment of value-maximising the firm. Therefore, a decline in performance prompts
managers to undertake risky M&As. Managers are also exposed to larger risks when they
engage in a risky M&A deal, motivating them to rationalise their motive by taking market-
anticipated risks and working hard to convince the market the performance will be boosted
through their superior management skills in risky projects. It is expected a firm’s current price
that deviates greatly from its 52-week highest stock price leads managers to undertake M&As
according to market anticipation. As a result, risky M&As would bring forth positive abnormal
returns for the firm. Managers engaging in risky acquisitions are also exposed in stricter
shareholder monitoring, requiring them to make greater efforts on deal negotiation, reflected
in the lower offer premiums they pay for the target. In a further analysis, the quality of the
decision that is based on market anticipation is assessed. Since the integration process of M&As
requires time, it is suggested that assessing the short-term performance may not fully reflect
the quality of a risky M&A decision. Therefore, the long-term performance of the deal is
studied to observe whether bidder managers make great efforts to improve the deal quality.
Chapter 1. Introduction
14
M&As form is a good platform to test managerial risk-seeking behaviour. First, they are a
major investment activity that could fundamentally alter the wealth status of the shareholders,
since small magnitude of change in wealth status does not touch investors’ risk appetizer
(Brenner, 1983). Further, it is common for shareholders to compensate managers engaging in
risky investment projects (Grinstein and Hribar, 2004; Graham et al. 2013). The feeling of loss,
part of human nature, tends to motivate the manager to tolerate more risks and accordingly
undertake risky M&As. Finally, M&As allow the researcher of this thesis to further explore
whether managers can control risks with their skills and efforts. It was also found that risk-
taking is distinguished from gambling insofar as it depends purely on probability.
Chapter 5 makes two main findings by studying a sample of 2,018 U.S. public acquisitions
announced between 1985 and 2014. First, the bidder reference point and target risks are
positively related, implying that when bidders perceive losses relating to the reference point
they are likely to undertake risky investment projects. Second, bidders engaging in risky
projects anticipated by the market tend to perform better, suggesting that managers who follow
market-anticipated risks will be rewarded. Third, it was also established that risky M&As
expose managers to strict shareholder monitoring, which motivates them to work hard.
Chapter 5 makes three distinct contributions to M&A literature. First, this is the first paper to
test the reference point effect on managerial risk-taking behaviour in the M&A context. It is
also suggested that human-related biases have a direct impact on investment behaviour. Second,
the researcher contributes to behaviour finance literature by documenting the evidence of the
reference point effect on both individual and institutional investors. Third, a contribution is
made to M&A literature, affirming that taking risks does not necessarily lead to negative
Chapter 1. Introduction
15
market reactions and taking risks according to market’s anticipation is an important source of
positive market reactions.
The remainder of this thesis is organised as follows: Chapter 2 provides a summary of relevant
literature on M&A motives, M&A processes, M&A performance and prospect theory. Chapter
3 studies the reference point theory in domestic and cross-border M&As in the U.K. market.
Chapter 4 examines the reference point effect on M&A misvaluations, while Chapter 5 assesses
the reference point effect on managerial risk-taking behaviour. Chapter 6 concludes this thesis.
16
CHAPTER 2: LITERATURE REVIEW
Chapter 2. Literature Review: M&A Motives
17
2.1. M&A motives
Researchers on mergers and acquisitions (M&As)2 focus on the causes and consequences.
M&As are one of the major investment activities undertaken by firms to create wealth for
shareholders, expand the size, or integrate resources for example. Prior studies have offered
various explanations for M&A motives. So as to prescribe corresponding strategies for
management following M&As, Trautwein (1990) recognised seven M&A motives, namely
efficiency theory, monopoly theory, valuation theory, empire-building theory, raider theory,
process theory and disturbance theory. Of these theories, he finds that the valuation theory,
empire-building theory and process theory play a more active role in M&A motives than the
efficiency theory and monopoly theory, whereas raider theory and disturbance theory have the
least explanatory power. Despite this, his study implies that no single theory can fully account
for an M&A motive.
Studies have also provided empirical evidence in explaining the M&A motives. Berkovitch
and Narayanan (1993) were the first to test the fitness of the three major M&A motives: synergy,
agency and hubris. The authors studied a sample of tender offers in the U.S. market between
1963 and 1988 and used the correlation between the gains of the target and the combination to
examine M&A motives. They assumed that the positive correlation between gains of the target
and the combination represented synergy, the negative correlation represented agency whilst
the zero correlation represented hubris. Their study indicates that about 75 percent of M&As
are motivated by synergy and the negative correlations are driven by agency, while hubris co-
exists with synergy. Hodgkinson and Partington (2008) put forward their idea in an alternative
market. They examined a sample of 529 successful M&A deals between 1984 and 1998 in the
2 In this thesis, the terms ‘takeovers’ and ‘mergers and acquisitions’ were used interchangeably, ‘mergers’ and
‘acquisitions’ also refer to M&As in some circumstances unless they are specified in the correspondent sessions.
Chapter 2. Literature Review: M&A Motives
18
U.K. market and found that synergy was the most plausible M&A motive and there is also
evidence for both agency and hubris. Raj and Forsyth (2003), who focused on the performance
of hubris-infected bidder managers, noted significant loss on the bidder announcement returns,
suggesting that one of the negative sources of bidder performance is managers’ overestimation
of M&A synergies.
Seth et al. (2000), who studied the bidders’ motives in the context of cross-border M&As,
affirm that synergy remains a primary motive for cross-border M&As. Studying 100
acquisitions of U.S. targets between 1981 and 1990, the authors found synergy is the
dominating M&A motive for domestic M&As consistent with the findings of Berkovitch and
Narayanan (1993). In addition, the total gains in cross-border M&As are 7.6 percent, which is
similar to the figure achieved in the study conducted by Bradly et al. (1988)3. Seth et al. (2002)
identified multiple sources of value creation and value destruction for cross-border M&As.
Based on their sample, asset sharing, reverse internationalisation of valuable intangible assets
and financial diversification are the main sources of value creation while risk reduction is the
only source for value destruction. Their paper suggests that synergies play an essential role in
cross-border M&As.
2.1.1. Synergies
The synergy hypothesis of M&As asserts that the value of combined firms after a merger is
larger than the sum of the returns generated by the two separated firms. Bradley et al. (1983)
found some support for this hypothesis, suggesting that M&As create synergies for the
combined entity by reallocating the resources of the firms. Unlike the information hypothesis,
which posits that managers have private information about the true value of the firm, authors
3 Bradly et al. (1988) found a 7.4 percent increase of the combined firm in domestic M&A sample.
Chapter 2. Literature Review: M&A Motives
19
suggest that managers generate gains by exploring potential synergies via M&As, which is
consistent with the synergy hypothesis. Bradly et al. (1988) investigated a sample of 236
successful tender offers in the period 1963 to 1984 and recorded an overall increase of 7.4
percent in the value of combined firms, providing additional support to the synergy hypothesis.
Jensen and Ruback (1983), by reviewing prior studies, found that M&As on average generate
gains, and targets benefit, whereas bidder firms at least do not lose.
A large body of M&A studies have summarised three main sources of synergies: operations,
finances and management improvement. Devos et al. (2009) attribute synergies to an
improvement of resources allocation. Using the value line forecasts to estimate synergy4, the
authors found that the average synergy for a sample of 246 large mergers increased the value
of the combined entity to 10.03 percent. Of particular note, operating synergies take up 8.38
percent and financial synergies account for the remaining part. Hoberg and Phillips (2010)
suggest that large synergies are obtained through acquisitions of targets whose products are
different from bidder firms’ rivals, suggesting that synergies are created by a product
differentiation strategy that intensifies bidders’ competitiveness and sets obstacles for their
rivals, making them hard to imitate. Using a large unique patent-merger dataset between 1984
and 2006, Bena and Li (2014) found synergies to be created through innovation activities
following M&As. Houston et al. (2001), who investigated a sample of the largest bank mergers
between 1985 and 1996, noted that bank mergers create value for bidders. In particular, the
authors identified the source of synergy in bank M&As as being mainly created from cost
savings rather than revenue enhancement.
4 In Devos et al. (2009), merger synergy was calculated as the forecasted incremental cash flows of the combined
firms relative to the sum of the premerger forecasted cash flows of bidding and target firms.
Chapter 2. Literature Review: M&A Motives
20
Lang et al. (1989) explained synergy gains with the Q theory. Tobin’s Q is defined as the ratio
of a firm’s market value to the replacement costs of its assets. Q ratio measures the firm’s
performance, and the well-performing firms are those with a high Q ratio whereas the poor-
performing firms are those with a low Q ratio. Acquisitions involve a high Q bidder and a low
Q target generating large total gains for the bidder, the target and the combined firm. The results
imply that managers of a well-performing firm have abundant resources to improve the firm’s
performance whereas a poor-performing firm lacking investment opportunities is an ideal
target for synergy exploitation. Servaes (1991) reported supportive evidence by applying Lang
et al.’s idea (1989) to a larger sample with additional controls.
M&As play a disciplinary role with regard to managers who perform poorly. Jensen and
Ruback (1983) indicate that the market for corporate control creates value for the firm.
Managers compete for the rights to manage corporate resources, hence, poor management is
removed while good management serves the best interests of the shareholders via M&As.
Jensen and Ruback’s view (1983) suggests that M&As enhance the firm’s efficiency by
removing an inefficiency management team. Palepu (1986) studied the inefficient management
effect on the likelihood of the firm to be acquired, and measured an inefficient management
team by means of average excess returns of the firm. His results show a negative relationship
between firms that are likely to be the M&A target and inefficient management, which is
significant at a 5% level. The conclusion of this inefficiency management hypothesis is that
managers who fail to ensure the commitment of value-maximisation for the firm are likely to
be replaced. Accordingly, Mitchell and Lehn (1990) found that firms whose managers perform
poorly following a merger is likely to be acquired by another firm. Martin and McConnell
(1991) affirmed high turnover rate for target managers following M&A completion and the
performance of these firms is well below that of the industry average prior to the M&A. In a
Chapter 2. Literature Review: M&A Motives
21
similar vein, Lehn and Zhao (2006), by studying 714 acquisitions during 1990 to 1998, found
that 47% of top managers of bidder firms are dismissed within 5 years (including 16% due to
takeovers). Therefore, takeover threats for managers tend to create synergy for the firm.
However, Agrawal and Jaffe (2003), who included a sample of over 2,000 M&As from 1926
to 1996, reported little evidence that inefficient firms will necessarily be acquired. Measuring
the operating performance and stock market returns for the target firms, the authors revealed
that takeover targets are not all those underperforming firms prior to the M&A announcement.
In addition, targets do not underperform in the decade prior to M&A announcements regardless
of whether the performance was measured with the firm’s operating performance or stock
market performance. Results are robust when the authors take account of the size of the firm,
industry and past performance of the targets for the targets’ operating returns, and the size,
book-to-market value and past returns for the stock market performance. Results continue to
hold when employing an alternative benchmark model to calculate targets’ stock returns. The
authors proposed two main reasons why their findings fail to support Palepu’s inefficient
management hypothesis (1986). First, the primary M&A motive is not to remove the inefficient
management team in the sample, as there is only a small proportion of targets perform
inefficiently prior to M&A announcement date. Second, their sample did not include
disciplinary or attempted M&As that support the inefficient management hypothesis.
Jensen (1988) suggests that the market for corporate control brings forth economic benefits for
shareholders, society and the corporate form of organisation. Moreover, takeovers bring about
major changes fitting for the future development of firms. Such fundamental changes including
new recruitment and resources, offer new top-managers opportunities, forsaking the old
strategies and facilities that are no longer beneficial to the firm. Further, the market for
Chapter 2. Literature Review: M&A Motives
22
corporate control effectively reallocates resources. Jensen (1988) indicates that major
restructuring activities create economic efficiency due to the influences of political and
economic conditions in the 1980s when many U.S. firms experienced revenue slowdown and
used M&As to cope with market changes. The author holds that the pressures of managers’
turnover and the political activity weaken M&A efficiency.
2.1.2. Agency problems
The firm is a separation of ownership and control, in which shareholders are the owners of the
firm (i.e. the principal) and managers are the agent hired by shareholders to manage the firm
(i.e. the agency). According to Jensen and Meckling (1976), this principal-agency relationship
is a financing contract between shareholders and managers, in which managers are committed
to maximising the wealth of shareholders. However, such a contract is violated when managers
pursue their own interests at the expenses of their shareholders (i.e. agency problems).
M&A studies have found evidence of agency problems in firms with sizable free cash flows.
Jensen (1986) suggests in the free cash flow hypothesis that managers of cash-rich firms tend
to distribute the excess of cash in unnecessary investment projects rather than paying out to
their shareholders. Harford (1999) provided empirical evidence in support of Jensen’s view.
Studying a large sample of U.S. M&As from 1950 to 1994, Harford found that managers of
cash-rich firms engaging in M&As due to low managerial ownership is where agency problems
are likely to occur. Cash-rich bidders receive negative market reactions and experience declines
in operating performance following M&As.
Contrary to the free cash flow hypothesis, Gregory (2005), focusing on the long-term
performance of U.K. market, noted that firms with abundant cash flows free of use outperform
Chapter 2. Literature Review: M&A Motives
23
those possessing low free cash flows. The author argued that firms with low free cash flows
are more likely to be financially constrained in the long term, and have difficulty in boosting
its performance following a merger. Managers who hold large free cash flows and whose firms
have a low Q ratio are likely to create value for the firm, in that the initial undervaluation tends
to be corrected by the market in the long term. Likewise, Lin and Lin (2014), using a sample
of Australia takeovers and measuring free cash flows with two proxies: excess cash holding
and excess accounting cash flow5, found that managers of cash-rich firms do not undertake
value-destroying M&As.
Morck et al. (1990) found “bad managers”6 who raise severe agency conflicts in M&As. Using
a sample of 326 U.S. M&As from 1975 to 1987, the authors found evidence of agency problems
through three types of value-destroying deals. First, agency problems are raised when managers
act for the purposes of job security and risk reduction of their own stock holdings. As such,
managers tend to undertake diversifying M&As associated with negative market reactions due
to unfamiliarity with an unrelated industry. Further, it is also suggested that bidders are likely
to overpay for high-growth targets to increase their personal benefits. Finally, firms performing
poorly in the past are likely to initiate bad deals. Harford et al. (2012) found entrenched
managers were less likely to pay for acquisitions with all-equity offers, avoiding creating larger
block-holders that diminish their power in the firm. Their M&A selection avoids acquiring
private targets that could create value for the firm7, and more importantly, they overpay for
targets whose synergy-generating ability is low.
5 Excess accounting cash flow was defined as the ratio of earnings after interest paid, tax paid and dividend paid
before depreciation to total assets. 6 “Bad managers” here refers to those who do not serve the best interests of their shareholders. 7 In Fuller et al. (2002), bidders’ acquisitions of private firms tend to generate positive announcement returns
while acquisitions of public firms generate negative announcement returns. Specifically, it is found that there is
negative 1 percent announcement returns when targets are public firms and positive 2.1 percent when targets are
private firms. In addition, Fuller et al. (2002) showed that offer premiums to private targets are lower than public
targets, which is also consistent with the finding in a study of French M&As over the period of 1966 and 1982 by
Eckbo and Longohr (1989).
Chapter 2. Literature Review: M&A Motives
24
Studies have shown one of the main sources for agency problems is the compensation
rewarded to top managers engaging in M&As. Studying a large sample of U.S. M&As, Harford
and Li (2007) found it was a great incentive for a bidder CEO to undertake an acquisition as
CEOs’ pay increases substantially after an acquisition regardless of the wealth of shareholders.
This is also supported by Guest (2009) who studied a comprehensive U.K. sample between
1984 and 2001. He established that CEOs’ salaries increase within a year after an acquisition
even though takeovers destroy the firm’s value.
Grinstein and Hribar (2004) revealed that it is managerial power that influences compensations
but not the deal performance, offering little evidence that better M&A performance increases
pay to the managers. Their study showed that 39% of companies reward their CEOs involved
in a completed deal due to the efforts the CEOs made, where efforts of those in the important
positions are easily to be seen. In contrast, Datta et al. (2001) and Falato (2008) found that
CEOs’ compensation is positively related to stock performance, indicating that compensation
boosts the firm’s performance, implying that there is low probability that compensation causes
agency problems.
Studies have also documented that not only bidder managers but also target managers could
cause agency problems, reflected in that target managers exchange private benefits such as
compensation or positions in the bidder firms following a merger for lower M&A offer
premiums. Analysing 311 large U.S. firms between 1995 and 1997, Hartzell et al. (2004) found
that target managers are likely to persuade shareholders to give up their control of a firm when
they are compensated with financial rewards or attractive positions in the bidder firms.
Chapter 2. Literature Review: M&A Motives
25
Likewise, Wulf (2004) in a sample of “mergers of equals”8 in the U.S. market between 1991
and 1999 found that target managers exchange shared governance in the bidder firms for lower
M&A offer premiums, which destroys the wealth of target shareholders, leading to 5.6 percent
lower than average abnormal returns. However, Bargeron et al. (2010) argued that target
managers’ retention is not as the result of lower M&A premiums. Their findings contradict
those of Hartzell et al. (2004) and Wulf (2004), indicating that bidders retain targets’ managers
who have skills and experience in the new combined firm to increase its value.
2.1.3. Behavioural finance theories
Behavioural finance has received a great deal of attention in M&As over the last three decades.
There are two threads in the literature that link behavioural finance with M&As, according to
a survey conducted by Baker and Wurgler (2012). The first thread of literature assumes that
the market is efficient while the managers are less than fully rational, in that managers cannot
benefit shareholders via M&As because there are no mispricing opportunities waiting to be
explored., and the hubris management hypothesis and the managerial overconfidence
hypothesis are the two of the most important hypotheses built upon this assumption. The second
thread of literature assumes that the market is inefficient while the managers are fully rational,
in that managers can explore market mispricing, and the over-reaction (or under-reaction)
hypothesis and the market-timing hypothesis are the two of the most important hypotheses built
upon this assumption.
8 According to Wulf (2004), “mergers of equals” was defined as that the bidder firm and the target firm is similar
in size.
Chapter 2. Literature Review: M&A Motives
26
2.1.3.1. Managerial overconfidence
Roll’s study (1986) was the first on managerial hubris in the context of M&As. The author
proposed his hubris management hypothesis as managers overpaying for targets due to an
overestimation to their own profit-generating ability. By reviewing U.S. and U.K. studies, Roll
(1986) identified overconfident managers and discussed them from three perspectives. First,
bidder managers are subject to hubris bias, in that the firm’s price falls after that a bid is
announced or fails at a later stage. Second, hubris infects target managers, as a target firm’s
price increases by 7 percent on average when a bid is announced and falls significantly when
the bid is withdrawn. Third, the performance of the combination entity is also affected by hubris
as it earns significantly negative abnormal returns following a merger. The hubris management
hypothesis provides evidence that an individual decision maker may not be rational, which
answers the question of why managers tend to pay for a takeover target with a price that is
higher than its market price.
Following hubris management hypothesis of Roll (1986), Hayward and Hambrick (1997), who
studied a sample of 106 large U.S. acquisitions, found that bidder CEOs paying too much for
large acquisitions is due to hubris bias. The authors identified three important sources of
managerial hubris: the firm’s recent performance, the recent media praise for CEOs’
performance and the CEOs’ self-importance. The authors suggest that any recent success of
the firm would enhance the confidence of the bidder managers. As a result, they overpay for
the targets. The media praise for CEOs also fosters CEOs’ overconfidence in investment
activities. Moreover, managers receiving with higher-than-average compensation and salary
tend to feel important in major investment decisions, which will make them believe their
abilities could improve the firm, thus overpaying for the targets. Similar to those of Roll (1986),
Hayward and Hambrick found hubris results in losses on bidder abnormal returns.
Chapter 2. Literature Review: M&A Motives
27
Malmendier and Tate (2005) revealed CEO overconfidence in a sample of 477 publicly traded
U.S. firms from 1980 to 1994. Measuring managerial overconfidence in the exercise of
managerial options and CEO stock holdings, the authors found that overconfident CEOs tend
to exercise their stock options at a later period and hold losers for a longer period. It is because
that they are over-optimistic about their firm’s prospects or have strong affiliation with their
own stocks, which miss the correct market timing for investments. Besides, their study also
revealed that CEOs whose firms have a higher investment to free cash flow ratio tend to
undertake more investments, suggesting that firms with overconfident managers are likely to
have sufficient internal funds. Using the same dataset as in the Malmendier and Tate’s work
(2005) and adopting some press-based proxies for managerial overconfidence, Malmendier and
Tate (2008) found that the market reacts negatively to a bid announced by overconfident CEOs
and affirmed that managerial overconfidence encourages more acquisitions since
overconfident CEOs are more aggressive than non-overconfident CEOs. These two related
studies lead the researcher to believe that small firms are less vulnerable to managerial
overconfidence while large firms tend to make more value-destroying deals due to
overconfidence.
Many studies have addressed the fundamental reason for managerial overconfidence. One
thread in M&A studies attributes managerial overconfidence to the market’s tendency to
explore the past best performance of big firms. Rau and Vermaelen (1998) put forward the
over-extrapolation hypothesis that the financial market and the managers tend to over-
extrapolate the past performance of the bidder and react more positively to those with low
book-to-market ratio (i.e. glamour firms) than firms with high book-to-market ratio (i.e. value
firms). The market’s reactions reinforce the confidence of the managers of glamour firms when
undertaking M&As. In addition, the authors found glamour firms perform better than value
Chapter 2. Literature Review: M&A Motives
28
firms in the short term as the market tends to reward firms with better past performance.
Making adjustments for the size of the firm and book-to-market value to estimate the long-term
performance of the bidder, the authors found that glamour bidders underperform value bidders.
They interpreted this phenomenon as that short-term momentum created by the market’s over-
extrapolation leads to correspondent long-term reversals when the market can correct its initial
reactions, and such a short-term momentum is stronger in glamour bidders than value bidders.
Sudarsanam and Mahate (2003) found consistent results when putting forward Rau and
Vermaelen’s idea (1998) for the U.K. market. They employed the price-to-earnings ratio (P/E)
and market-to-book value (MTBV) for glamour and value bidders and calculated shareholder
returns with a range of benchmark models. Their findings suggest that a lower P/E bidder (a
value bidder) outperforms a higher P/E bidder (a glamour bidder) and a lower MTBV firm
outperforms a higher MTBV firm in the long term.
Another thread in relevant literature has largely attributed managerial overconfidence to
managerial self-attribution bias. Attribution bias was first discussed in Heider’s attribution
theory (1958) in which suggests that people tend to make systematic errors and attempt to
attribute those to external factors. Doukas and Petmezas (2007) tested this theory on an M&A
testing ground. Studying a sample of 5,334 successful U.K. private acquisitions between 1980
and 2004, the authors recorded that bidders in the higher order acquisition9 underperform those
in the lower order acquisition. They describe this phenomenon as the self-attribution bias of
managers who believe initial success is due to their own superior ability and so undertake more
M&A deals, whilst managers attribute negative market reactions of the subsequent deals as bad
luck. Billet and Qian (2008) provided some similar findings by focusing on the U.S. market.
9 According to Doukas and Petmezas (2007), higher order acquisitions refer to bidders who undertake five or more
deals within a three-year period.
Chapter 2. Literature Review: M&A Motives
29
After controlling for the order of the deal, their findings showed that the first bids have higher
abnormal returns whilst the higher order bids generate negative abnormal returns. In addition,
bidders who have experience of undertaking successful acquisitions before are likely to engage
in more value-destroying M&A deals. It is suggested that bidder managers are cautious in their
first bids but are affected by self-attribution bias when undertaking subsequent bids.
However, Aktas et al. (2009) explain the findings of Doukas and Petmezas (2007) and Billet
and Qian (2008) as the learning hypothesis. They suggest that the negative returns to the bidder
firm during the time of a bid announcement are a result of the learning behaviour of bidder
managers. The authors reason that bidder managers who undertake multiple bids tend to learn
lessons from the previous deal and become more risk-averse for subsequent deals, which leads
to negative abnormal returns. This study links a positive relationship between market reactions
and managerial risk-taking. It is worth noting that the learning hypothesis weakens in the case
of a manager not being able to engage in multiple bids prior to the current bid during his or her
professional career, hence some of them do not have a chance to learn from their experience,
notably young managers.
In addition to the extrapolation hypothesis and the self-attribution bias, many additional studies
have also explained causes and consequences of managerial overconfidence in the M&A
context. Moeller et al. (2004) affirm that bidder size plays an important role in managerial
overconfidence, suggesting that larger firms are likely to be overconfident as managers have
great discretionary power over the firms, resulting in fewer obstacles allocating a firm’s
resources. Since acquiring public firms is the quickest way of size expansion for a firm,
managers of large firms are likely to overpay for public targets with high growth opportunities.
Andreou et al. (2016) found a significant negative relationship between managerial
Chapter 2. Literature Review: M&A Motives
30
overconfidence and the value creation of corporate diversification. Analysing a sample of
diversification M&As, the authors found overconfident managers destroy firm value by 12.4
to 14.1 percent as opposed to rational managers do.
Despite the main stream of the literature that links managerial overconfidence and corporate
investments, which suggests that managers who overestimate their own ability make value-
destroying deals, a few papers note the bright side of managerial overconfidence. For example,
Galasso and Simcoe (2010) found a positive relation between managerial overconfidence and
a firm’s innovative performance. Using a CEO stock-option exercise for managerial
overconfidence consistent with that used by Malmendier and Tate (2005), the authors
discovered that overconfident managers tend to boost the innovative performance of a firm, as
overconfidence leads the managers to underestimate the probability of the failure of deals,
encouraging them to pursue innovative activities. In addition, overconfident managers are
likely to make significant changes in innovative strategies that make the firm stand out in a
competitive environment, which is more pronounced in competitive industries. Hirshleifer et
al. (2012) who measured managerial overconfidence with options and press-based proxies for
a sample of U.S. CEOs during 1993 to 2003 also found that overconfidence drives managers
to engage in riskier projects, greater investment in innovative activities and greater innovative
activities, suggesting that overconfident managers are better innovators and more likely to be
hired by high growth industries.
2.1.3.2. Investors’ overreactions and underreactions
Debondt and Thaler (1985) postulate the overreaction hypothesis that news-oriented investors
exhibit reactions driven by unexpected and dramatic news events, which lead to market
fluctuations and make it predictable. Using the monthly data of U.S. stock returns, the authors
Chapter 2. Literature Review: M&A Motives
31
uncovered evidence of inefficiency in a weak form market, thus proposing two testable
hypotheses. First, price reverses follow extreme price movements that have occurred over the
previous days. Second, such a price adjustment tends to be greater when the initial price
movement is dramatic. Their study reveals that portfolios that have lower returns in the past 5
years outperform those have higher returns by 25 percent, suggesting that the market is sluggish
to respond to fresh information and overreaction is due to mean-reversion.
Consistent with Debondt and Thaler (1985), Barberis and Vishny (1998) suggest that the
overreaction or underreaction is due to investors’ mistakes. When a company announces good
news, the market tends to be over-optimistic about the firm’s prospects, which pushes the
firm’s price to high levels. However, the price was revised when investors recognise the
fundamental reason for the price boost was due to overreactions. In a similar vein, investors
also revise their pessimism about loser stocks in the long term.
The overreaction hypothesis has been challenged by many researchers. Jordan and Pettengill
(1990), who duplicated Debondt and Thaler’s study (1985) in a different sample period, found
that losers become winners but winners do not lose. In addition, Fama and French (1996)
proposed a three-factor model to account for reversal of long-term returns as they argued size
effect on a firm’s abnormal returns, affirming that small firms are associated with higher risks,
while large firms are associated with lower risks. It became evident that small firms are likely
to generate higher returns than larger firms.
It is generally believed that evidence regarding the underration hypothesis is much robust than
that regarding the overreaction hypothesis. Barberis and Vishny (1998) documented the
mechanism for investors’ underreactions. When a company performs well (badly) the market
Chapter 2. Literature Review: M&A Motives
32
tends to raise (lower) its stock price and this will be further raised (lowered) when the
performance of the firm is well supported by its earnings performance in a subsequent period,
which often occurs in a short period of time (3 to 5 months) compared with the overreaction
(3 to 5 years). Whilst theorists who suggest high risks are associated with high returns, Barberis
and Vishny (1998) offered plausible explanations from a psychological perspective on how
investors make mistakes when processing new information, suggesting that managers can
formulate profitable investment strategies by taking full advantage of such mistakes. Edwards
(1968) suggests that conservatism bias of investors, suggesting that underreaction is formed
due to slow updating belief. In the light of this, Li and Yu (2012) found that conservative
investors focus on the market 52-week high and historical high, affirming the investors
attention hypothesis.
2.1.3.3. The misvaluation hypothesis
Shleifer and Vishny (2003) developed a model of M&As based on the assumption that
managers are rational and the market is less than fully rational. The model explains ‘who
acquires whom, the choice of the medium of payment, the valuation consequences of mergers
and merger waves’ (Shleifer and Vishny, 2003: 295). Their study focuses on the limitations of
the neo-classical theory of M&As, which focuses on industry-specific shocks, and the effect of
the method of payment on M&As, which is not clearly interpreted. In addition, reasons why
cash-financed M&As generate long-term positive earnings while stock-financed M&As
generate long-term negative earnings have not been well explained. So as to explore these
reasons, their study presents a simple valuation model of M&As to explain the short- and long-
term gains of the bidder. The model implies that the bidder can produce a positive synergy in
the short term yet few synergies in the long term. Important implications drawn from this model
are presented as follows:
Chapter 2. Literature Review: M&A Motives
33
Suppose the target firm is 0 and the bidder firm is 1. Stock volumes for firms 0 and 1 are K and
K1, and price per unit is Q and Q1, and Q1 being greater than Q. The price per unit for the
combined equity is S, q is the long-term price. Bidders would pay a price for the target, denoted
P. If synergies exist, P lies between Q and S (i.e. Q<P<S).
1. The combined market value is S (K+K1)-K1Q1-KQ, which is the value of the new combined
firms minus the value of each individual firms involved in an M&A deal.
2. The target value in the short term is (P-Q) K, suggesting that the offer price is over market
value.
3. The bidder value is made up of two parts, one part is the gains the bidders earned as long as
their pay is less than S, the other is the loss of value dilution of the bidder firm’s capital, from
Q1 to S or (S-P) K+(S-Q1) K.
4. Given that there is no long-term profit for the combined firm when acquisitions are funded
with cash since the gains to the target firm is the loss on the bidder firm, the effect on the value
of the target is K (P-q), whereas the value of the bidder is K (q-P).
5. For stock-financed acquisitions, suppose x is the ratio of deal value over combined firm
value or x = PK/S (K+K1). In the long term, this share is xq (K+K1) = q (P/S) K;
6. The net long-term gains to target shareholders in stock-financed acquisitions are q (P/S) K-
qK = qK (P/S-1), where qK is the stand-alone value of target in the long term, whereas for the
bidder shareholders, it is qK (1-P/S).
Chapter 2. Literature Review: M&A Motives
34
Shleifer and Vishny’s misvaluation model (2003) indicates that bidders are those with a
considerable valuation whereas targets are those with a low valuation. Further, the long-term
effect of cash-financed acquisitions suggests that the target is undervalued whereas the net
long-term effect of a stock deal generates a short-term loss and long-term gains, explaining
why an overvalued bidder tends to use stocks as ‘cheap currency’ to pay for targets. Moreover,
there is a reverse effect between the bidder firm and the target firm, i.e. what the bidder loses
is the target gains. Finally, the model reveals that both the target and the bidder firms could
gain in the short term if there is synergy from the mergers.
Shleifer and Vishny (2003) also reason that M&As are a trade-off game between the bidder
and the target, since they involve a high valuation bidder and a low valuation target based on
the rationale that a low valuation firm may find it hard to create synergy through the acquisition
of a high valuation firm. Their study suggests that a bidder buys a target to create long-term
value whereas the target focuses on short-term gains; hence, targets voluntarily accept
overvalued stocks for a cash-out purpose. All of these imply that bidder managers attempt to
maximise the interest of shareholders through exploring the market’s mispricing in an
inefficient market (Shleifer and Vishny, 2003: 296).
Therefore, Shleifer and Vishny (2003) assume that bidder managers are rational whereas target
managers are irrational as since they accept overvalued stocks. However, this view has been
challenged by Rhodes-Kropf and Vismanthan. (2004) who contend that rational managers
accept overvalued stocks by mistake. According to this study, target managers are misled when
the market valuation is high, and accept overvalued stocks as they overestimate synergies.
Further, Rhodes-Kropf et al. (2005) identified three types of specific valuation errors that drive
merger waves. By decomposing the conventional MTBV proxy into three components: the
Chapter 2. Literature Review: M&A Motives
35
firm-specific deviations from short-run industry pricing (firm-level misvaluation), sector-
factor, short-run deviations from a firm’s long-run mispricing (sector-level misvaluation), and
long-run pricing to book (long-run growth opportunities), the authors found that high MTBV
firms buy low MTBV firms, which is attributed to firm- and sector-level valuation errors. Firms
of low long-run growth opportunities tend to buy firms with high long-run growth opportunities.
Their study suggests a positive correlation between stock-financed acquisitions and market-
wide valuation.
In this connection, Dong et al. (2006) affirm that the misvaluation hypothesis is in favour of
high valuation periods. Analysing a sample of U.S. M&As from 1978 to 2000, the authors
claim that evidence for the misvaluation hypothesis is more relevant for M&As after 1990 than
before, whereas the evidence for the Q hypothesis is more relevant before 1990 than after.
Their findings suggest that the misvaluation hypothesis reconciles the case that the market-
wide valuation is high whereas the Q hypothesis is employed to explain how well-run bidders
explore synergies from poorly-performed targets. In addition, the authors find that more
overvalued bidders prefer to pay for targets with stocks rather than cash, choose mergers rather
than tender offers, and pay higher offer premiums. Their results are consistent with those
predicted by Shleifer and Vishny (2003).
Dong et al. (2006) investigated the effectiveness of the Q hypothesis and the misvaluation
hypothesis in different sample periods but fail to address the concern as to whether bidders
paying with overvalued stocks are being rational. According to Shleifer and Vishny (2003),
bidders who use overvalued stocks as a means of payment for M&As receive negative market
reaction in the short run, leading shareholders to believe that the firm does not generate
synergies. Their real purpose of using overvalued stocks is to dilute overvaluation and create
Chapter 2. Literature Review: M&A Motives
36
value for the firm, since holding stocks in an overvalued market incur risks for the firm (Jensen,
2005). In this respect, Ang and Cheng (2006) assessed the long-term performance of the bidder
who pays for acquisitions with stocks. Studying a sample of completed stock-financed
acquisitions, the authors find that bidders undertaking M&As with overvalued stocks perform
better than those engaging in M&As but not using overvalued stocks. Their study indicates that
not all M&As are driven by misvaluation, since their data show a small fraction of the stock-
financed acquisitions involve an undervalued bidder, which contrasts with Savour and Lu’s
view (2009) in their assumption that all stock-financed acquisitions are misvaluation-driven.
The misvaluation hypothesis has been challenged by many studies. Firstly, in line with Ang
and Cheng (2006), Fu et al. (2013) suggest that ‘The existence of relative overvaluation
between the bidder and the target is a necessary, but not a sufficient condition for the
acquisition to benefit acquirer shareholders’, as their M&A sample shows that two-thirds of
stock-financed acquisitions are motivated by a bidder firm’s overvaluation. They suggest that
bidder shareholders only benefit from overvalued stocks if low premiums are paid to the target
shareholders. Using a sample of 1,319 stock-financed acquisitions between 1985 and 2006, the
authors found that overvalued bidders do not outperform those firms in the control sample,
reflected in significantly worse operating performance and worse stock returns five years
following the M&As. By including corporate governance related proxies, their study reveals
severe agency problems in an overvalued firm, indicating that acquisitions driven by
misevaluation destroy a firm’s value when the firm experiences agency problems.
Further, Eckbo et al. (2016) suggest that ‘overvaluation reduces the all-stock payment
propensity’, which contradicts the idea of the misvaluation hypothesis. Using aggregate mutual
fund flows as an instrumental variable of the pricing error, the authors indicate that the shock
Chapter 2. Literature Review: M&A Motives
37
caused by mutual fund flows pushes up the market valuation but it does not increase the volume
of stock-financed acquisitions. Their study challenges the view that managers use overvalued
stocks to time the market. The authors find that bidders who pay with overvalued stocks for
financing acquisitions tend to be small firms and non-dividend paying firms with low leverage,
indicating that firms that are financially constrained avoid using cash as a means of payment
for financing M&As.
This section has reviewed three broad types of M&A motives. According to these studies
reviewed, there is no single theory that can fully address the M&A motive, since it varies
significantly in any particular sample at any particular time, which calls for up-to-date evidence.
Jensen and Ruback (1983), by reviewing a quantity of M&A research up to the 1980s,
suggesting that the primary M&A motive is synergies, and the small role the human factors
played on M&As weakened prior to 1990. This is also supported by a number of M&A research
works (Bradley et al., 1983; Bradley et al., 1988; Jensen, 1988; Lang et al., 1989; Servaes,
1991), whilst Jensen (1986) indicates the agency problem of free cash flow where managers of
cash-rich firms invest unwisely, and Roll (1986) suggests hubris-infected managers engage in
value-destroying M&A deals. Shleifer and Vishny (2003) suggest that M&As are driven by the
relative valuation of the two firms involved, indicating the role of managers’ timing the market.
Dong et al. (2006), who studied the misvaluation hypothesis and the Q hypothesis in the context
of M&As, noted that synergy-driven acquisitions are more likely to be found prior to 1990
whereas the misvaluation-driven acquisitions are common after 1990.
Chapter 2. Literature Review: M&A Process
38
2.2. M&A process
2.2.1. Selecting a target
Jemison and Sitkin (1986), who explored the M&A motive and M&A process, identified three
important M&A motives for bidder management: strategic fit, organisational fit and acquisition
process. The first two motives indicate that the acquisition should be financially or strategically
fit for the firm’s prospects, while the latter one indicates the essential role of M&A participants,
including analysts and key managers, played during the course of M&As. In this line of
thinking, human factors have a substantial impact on the M&A process.
The issue concerning “who buys whom” in the M&A context has been discussed for years.
Takeover targets are largely predicted within the scope of the bidders’ M&A motives. The
synergy hypothesis predicts that the less efficient firms are likely to be a takeover target, as the
bidder views target’s price discounts as a source of synergies generation. Investigating the
probability of takeover targets, Papelu (1986) proposed six hypotheses relating to “who buys
whom”, including the inefficient management hypothesis, the growth-resource mismatch
hypothesis, the industry disturbance hypothesis, the size hypothesis, the market-to-book
hypothesis and the price-earnings hypothesis, identifying that firms with an inefficient
management, smaller size, mismatch between resource and growth opportunity10 and high
leverage ratio are probably the target firm in an M&A deal.
Brar et al. (2009) who examined the probability of takeover targets by extending Palepu’s
model (1986) with inclusions of price momentum, trading volume and a measure of market
sentiment for a large sample of European and cross-border acquisitions found that takeover
10 According to Palepu (1986), the mismatch between resource and growth opportunity refers to a rich resource
with low growth opportunity or poor-resources with high growth opportunity.
Chapter 2. Literature Review: M&A Process
39
targets are small, undervalued and less liquid relative to bidders. Their findings are in
accordance with Palepu (1986). In addition, the authors suggest that takeover targets are those
firms with low sales growth and strong short-term price momentum and their shares are
actively traded prior to the announcement date, indicating that arbitrage around the M&A
announcement date leads the price of the target to be hard to value. The authors also noted that
takeover targets are those firms with incompetent managers, inducing other managers to
explore the potential synergies through takeovers.
Jensen (1988) notes that managers who compete over the rights to control a firm’s resources
initiate M&As. Hence, M&As are a vehicle of integrating the bidder firms’ and the target firms’
resources, suggesting that poorly performing firms are likely to be acquired in M&As.
Similarly, Mitchel and Lehn (1990) found that firms whose managers made value-destroying
deals are likely to subsequently become takeover targets. Lang et al. (1989) and Servaes (1991)
also provided supportive evidence that inefficient firms are likely to be takeover targets. Their
findings suggest that acquisitions enhance value for the firm when they involve a high Q bidder
and a low Q target. The Q hypothesis indicates that high Q bidders are easier to explore
synergies from the poor-performing firms than from the well-performing firms through M&As.
Another line of literature suggests that it may not be the case that acquired firms are in a weak
position prior to an M&A. Bradly et al. (1983) documented that targets are those ‘sitting on the
gold mine’, suggesting that targets hold excess cash, hence, they are attractive to firms
experiencing financial constraints. Agrawal and Jeffe (2003) find direct evidence that targets
do not underperform prior to M&A announcements, and their results continue to hold
regardless of various measures employed to influence the firm’s performance. Rhodes-Kropf
and Robinson (2008) showed evidence of ‘like buys like’, suggesting M&As involve a firm
Chapter 2. Literature Review: M&A Process
40
with similar prospects. Based on the relative bargaining power of the bidder and the target,
their model suggests that bidders tend to acquire firms of a similar MTBV. Levis (2011)
suggests that firms of high revenue growth opportunities but high operating costs are probable
takeover targets whereas firms of lower growth opportunities but high cost efficiency are
probable takeover bidders. Investigating a sample of newly-listed firms that become takeover
targets shortly after their initial public offerings, De and Jindra (2012) noted firms that perform
well in operations or in the stock market are likely to be acquired due to superior post-IPO
performance.
Studies have addressed the issue concerning “who buys whom” from other angles. Massa and
Zhang (2009) found M&As are driven by the popularity difference between the two firms
involved. They identified ‘cosmetic mergers’ where a less popular firm tends to buy a more
popular firm. This investment style is more prevalent when the popularity difference 11
between the target and the bidder is significant. The popularity of an acquired firm creates a
‘hole’ effect for a less popular firm, increasing the probability that the market will re-evaluate
the bidder firm. Shleifer and Vishny (2003), however, suggest that the undervalued firm is
probably the takeover target as it provides great scope for value dilution for the highly valued
firm. If the primary M&A motive of the bidder is to eliminate any takeover threats, larger firms
in the same industry as the bidders are the probable targets since they can quickly help expand
the bidder firm’s size and reinforce their competitiveness among the industry peers (Gorton et
al., 2009). Indeed, Hoberg and Phillip (2010) showed that firms offering the bidder some hard-
to-imitate assets are attractive takeover targets.
11 The authors used data on the flows in mutual funds in different styles to construct a measure of popularity.
Chapter 2. Literature Review: M&A Process
41
Not only bidders look for targets, but also targets actively search for bidders (i.e. target-initiated
deals12). Studying a sample of over 3,800 U.S. public acquisitions between 1996 and 2009,
Eckbo et al. (2014) found 45% of acquisitions in their M&A sample were initiated by targets.
Their findings are also supported by Boone et al. (2007) who observed that about half of
takeovers of 400 U.S. takeovers announced between 1989 and 1999 had been initiated by
investment bankers hired by the target whose aim was to sell the firm. Masulis and Simsir
(2015) suggest a large number of M&As deals announced between 1997 and 2006 are target-
initiated: 237 as opposed to 408 bidder-initiated deals in their sample of 645 M&A deals. The
authors documented that target financial or economic weakness, target financial constraints and
economy-wide shocks were the key factors that determined target-initiated deals. The target-
initiated deals that are not strategically or financially fit for bidders reinforce bidder’s
bargaining power, resulting in lower levels of offer premiums, target announcement returns
and the deal value to EBITDA multiples of the target-initiated deals compared with the bidder-
initiated deals (Masulis and Simsir, 2015).
2.2.2. Paying for a target
Paying for a target is of first-order importance in M&As, since how much a bidder pays for
acquisitions directly determines the wealth distribution between the shareholders of the two
firms involved. Though a high offer premium increases the likelihood of the deal success, it is
at the expenses of the wealth of bidder shareholders. Because a low offer premium is hard to
persuade the target to give up the firm given that target shareholders are loss-averse (Baker et
al. 2012). However, the argument above leads the researcher to a question as to how much
offer premiums are too much, as on-one can give an accurate figure of the valuation of the firm
since valuing a target requires additional considerations regarding the decision makers. Two of
12 Information on who initiates a deal is not identified in SDC database.
Chapter 2. Literature Review: M&A Process
42
the most important factors influencing M&A valuation is how much M&A offer premiums the
bidder pays for the target, and by what means the bidder is willing to pay. The following section
reviews literature on the offer premium and the method of payment.
2.2.2.1. M&A offer premiums
Schwert (1996) explains offer premiums as the sum of the target price run-up and the mark-
up.13 The author found that final offer premiums increase with the pre-bid target stock price
run-up while the price mark-up fails to explain a majority of offer premiums. His study
measured the target price run-up with the target abnormal stock returns from 42 days to 1 day
prior to the takeover announcement date relative to the abnormal stock returns during the
announcement date, and offer premiums with target abnormal stock returns between 42 days
prior to the takeover announcement date and 126 days after takeovers.14 Analysing a sample of
1,814 U.S. mergers and tender offers between 1975 and 1991, the author established that for a
one dollar increases in target price run-up would cost a bidder firm an extra 1.13 dollar.
Similarly, Betton et al. (2008), who studied the relationship between the target price run-up
and initial offer premiums, found that for a dollar increase in the price run-up increased the
offer price by 0.8 dollars.
The minimum offer price equals the current market price of the target firm given that the firm’s
value is fairly estimated. It is commonly believed that the offer price should be higher than the
market price of the target, otherwise the target will refuse the offer, arguing that it will bring
forth synergies to the combined firm. However, in some cases firms are sold below market
value or with negative offer premiums. Weitzel and Kling (2014) suggest that targets
13 In other words, offer premiums were defined as the difference between abnormal returns and the price run-up. 14 For those targets delisted before 126 days after takeovers, he used the day when they were delisted.
Chapter 2. Literature Review: M&A Process
43
compensate bidders in the acquisitions on the rationale that they believe there will be post-
merger synergies. Secondly, targets compensate bidders who agree to accept their overvalued
shares, since small firms find it hard to justify overvaluation. Finally, targets who experience
severe shortage in liquidity tend to sell firms with negative offer premiums.
However, in most cases, the offer premium is positive and underlines the overpayment
hypothesis or the synergy hypothesis. If offer premiums are driven by wealth creation purpose,
bidder firms gain positive abnormal returns around the announcement date (i.e. the synergy
hypothesis). In agreement with this line of thinking, Antoniou et al. (2008) found that offer
premiums and bidder returns are positively related in a sample of 396 U.K. public acquisitions
announced between 1985 and 2004.
Evidence suggests lower offer premiums are paid to the target shareholders if the needs of
target managers are met. Wulf (2004) suggests bidders who guarantee the retention of target
managers in the combined firm tend to pay a lower offer premium for the target. Studying the
relationship between offer premiums and the choices of payment method in a sample of 2,959
acquisitions between 1983 and 2004, Zhang (2009) found that offer premiums decrease when
cash is paid as a result of increasing popularity of cash. Their findings suggest that cash
payments increase the bargaining power of the bidder when negotiating with targets who have
a strong cash preference. In this line of thinking, bidders, by offering what the target manager
wants, save takeover costs.
The overpayment hypothesis suggests that bidders tend to overbid for the target if the offer
premium is greater than expected synergies generated following the combined firm for a variety
of reasons. Roll (1986) indicates that bidders who overestimate the synergies of a combined
Chapter 2. Literature Review: M&A Process
44
firm or over-optimistic about their own management skills tend to overpay for targets. Rhodes-
Kropf and Viswanathan (2004) explained this as overestimation about the synergies increasing
with valuation errors. Following this line of discussion in relevant literature, Hayward and
Hambrick (1997) provided related support that hubris-infected managers tend to overpay for
acquisitions. Investigating offer premiums for a sample of large U.S. M&A deals, the authors
suggest that CEOs whose firms have performed well recently, who are praised by the media or
have a high salary have strong sense of self-importance, which increase high offer premiums
to the target shareholders. Antoniou et al. (2008) found overpayment for diversified M&A
deals and attribute this phenomenon to over-optimistic CEOs stepping into an unrelated field.
Baker et al. (2012) found confident bidders who pay according to the target 52-week high so
as to successfully obtain an M&A deal, reasoning this as the reference point theory of an M&A.
Since targets have a loss-aversion tendency, a price that is significantly lower than this
reference point price will increase the probability of the deal being rejected by the target. The
authors also measured the market reactions to the M&A pricing decisions that depend on the
reference point, and found that the payment according to the target reference point was
associated with negative market reactions, as the market regards this as overpayment.
Rather than studies relating overpayment to bidders’ overconfidence or reference dependence
bias, others explaining it with the bidders’ rationality. Jensen (2005) suggests that bidders
overbidding the target with overvalued equity is a measure to avoid the risks of holding those
stocks in an overvalued market. The firm’s value will be destroyed when the initial
overvaluation is corrected by the market. This view is also supported by the prediction of
Shleifer and Vishny’s theoretical model (2003), suggesting that bidder managers attempting to
dilute their overvaluation are likely to overpay for a less overvalued target. In order to reduce
the risks of stocks being down-priced by the market, bidders offer high M&A premiums to
Chapter 2. Literature Review: M&A Process
45
smooth the M&A process. Evidence that managers do not pay high offer premiums to the target
shareholders because of managerial irrationality is also supported by many additional studies
(Dong et al., 2006; Fu et al., 2013).
Studies have related overpayment to agency problems. Specifically, bidders’ overpayment is
due to lucrative compensation paid to managers who initiate or complete M&A deals. Harford
and Li (2007) and Guest (2009) suggest that CEOs engaging in an M&A deal are better off
regardless of the quality of the deal. Since the primary purpose for the manager is to obtain
rewards through M&As, they overpay for the target so as to smooth the M&A process. Jensen
(1986) also indicates that managers of cash-rich firms tend to pay excessively for these ill-
conceived investments.
It can therefore be concluded that offer premiums should bring forth positive abnormal returns
to the bidder if they are in line with the synergy hypothesis, while negative abnormal returns
will accrue to the bidder if they are explained with the overpayment hypothesis. In this pursuit,
Diaz et al. (2009) identified a quadratic relationship between the offer premium and bidder
abnormal returns, reasoning that the lower level of the offer premium represents its role of
synergies whereas the higher level represents overpayment.
2.2.2.2. Method of payment hypotheses
The choices of payment methods are another important factor influencing M&A valuation.
This section reviews four broad categories of explanations regarding the method of payment:
tax considerations, the signalling hypothesis, capital structure and corporate control and
behavioural finance, followed by relevant empirical evidence for each explanation.
Chapter 2. Literature Review: M&A Process
46
2.2.2.2.1. Tax considerations
Studies have linked the method of payment with corporate taxes, suggesting that targets would
demand high offer premiums when M&As were solely financed by cash. In accordance with
capital gains taxes, there will be an immediate tax on capital gains for cash while tax on capital
gains for stocks is deferred until gains are realised. Therefore, high offer premiums are
associated with cash as a compensation for the targets. Wansley et al. (1983) who studied the
tax effect on the choice of payment method found that target cumulative average abnormal
returns (CAARs), measured with 41 trading day stock returns after M&As, are 33.54 percent
when the means of payment for M&As is cash, but this figure nearly doubles when it is
substituted with stock payments. Likewise, studying a sample of 204 pairs of mergers between
1977 and 1982, Huang and Walking (1987) reported CAARs for cash payments, stock
payments and the combination of cash and stocks were 29.3 percent, 14.4 percent and 23.3
percent respectively.
By contrast, Franks et al. (1988) found that the presence of the all-cash offer premium effect
was pronounced before capital gain taxes were introduced into the U.K. market, suggesting
little evidence of the tax treatment effect on the method of payment. They examined a sample
of 2,500 acquisitions in both the U.K. and the U.S. markets between 1955 and 1985 and found
that U.K. firms had a much stronger cash preference than their U.S. counterparts. In addition,
the period of 1965 to 1969 saw a significant decline in cash-financed acquisitions as compared
with that in the period of 1960 to1964, before the capital tax gains law came into effect: 18.6
percent compared with 29.2 percent. Moreover, their study indicates that cash payments,
though associated with higher offer premiums than stock payments, reap more benefits for
long-term shareholders due to the valuation effect.
Chapter 2. Literature Review: M&A Process
47
It can be argued that early studies that examined the tax treatment effect on the choice of
payment methods based on the periods when the method of payment effect has received little
attention. In this line of thinking, examining the tax treatment effect on the method of payment
in a public U.S. acquisition sample between 1985 and 2004, Ismail and Krause (2010) provide
the most up-to-date evidence that fails to detect the tax treatment effect on the method of
payment. Their study reinforces Frank et al.’s findings (1988), suggesting the tax treatment
effect has little impact of on the method of payment.
2.2.2.2.2. Signalling hypothesis
Another classic explanation for the choice of payment methods is the signalling hypothesis
assuming that the market is transparent and there are no transaction costs and taxes, the method
of payment choices should be irrelevant to finance an M&A deal. However, Myers and Majuf
(1984) found evidence of information asymmetry, as managers know better than the market
about the value of a firm, in that they are able to choose the right sources of financing for
acquisitions. As such, stock payments signal to the market that the firm is overvalued and thus
incur negative market reactions, whilst a cash payment indicates that the firm is undervalued,
resulting in positive market reactions.
According to this literature, Travos (1987) found direct evidence that stock payments are
associated with negative bidder announcement returns whereas a cash payment is associated
with positive bidder announcement returns, claiming that stock payments convey bad news to
the market whereas as cash payment conveys good news to the market. Using a sample of
successful acquisitions between 1972 and 1982, the author examined a series of factors that
had a substantial impact on bidder announcement returns highlighted in the prior M&A
literature, including the method of payment, the offer premium, relative size and the type of
Chapter 2. Literature Review: M&A Process
48
mergers. Surprisingly, it is suggested that only the method of payment has a significant impact
on bidder announcement returns, and stock payments are negatively related to bidder
announcement returns, which is consistent with M&A literature.
Hansen (1987) regards stock payments as a risk-sharing agreement between the bidder and
target firms. With high levels of information uncertainty, bidders are likely to be misled by the
valuation of the target. So as to avoid overpayment, bidders are prone to pay for acquisitions
with stocks, requiring the targets to share the downward risk of stock overvaluation, while cash
is a clear cut between the bidder and the target and paying acquisitions with cash indicate
undervalued bidders. His study underscores the central role of the risk relating to the method
of payment, providing a rationale on why bidders choose stocks for acquisitions given that
abnormal returns are lower than those paid with cash.
Fishman (1989) claims that cash payments played a pre-emptive role in M&As involving
multiple bidders. He proposed a model of pre-emptive bidding, suggesting that cash payment
signals to the target that the bidder’s valuation is high whereas stock payments indicate the
opposite case. The role that cash played in enhancing the advantages of the bidder was also
addressed in Franks et al.’s U.K. study (1988). Cornu and Isakov (2000) found supportive
evidence that cash offers deter competition, working better than debt or equity. Consistent with
the signalling hypothesis, Cornu and Isakov (2000) reason that cash offers signal to the market
a high-valuing bidder who is capable of managing the deal.
In the presence of information uncertainty, the market perceives stock payments as
overvaluation by the bidder or that uncertainty regarding expected synergies increases with
stock payments. As such, stock payments incur negative market reactions whereas cash
Chapter 2. Literature Review: M&A Process
49
payment brings forth positive market reactions. However, a number of studies have challenged
this view, arguing that bidders do not underperform when financing an acquisition with stocks.
Savor and Lu (2009) found that bidders who are able to complete an M&A deal with stocks
would outperform those who fail to do so. Studying the joint effects of seasonal equity offerings
(SEOs) and method of payment on bidder gains, Golubov et al. (2015) found stock-financed
acquisitions are not the source of value-destruction of an M&A deal.
2.2.2.2.3. Capital structure and corporate control
The choices of payment methods also relate to a firm’s capital structure decision. Pecking order
theory suggests that firms prefer to use internal resources to fund investments and will not use
external resources until their internal resources are exhausted. The method of payment is the
means that the bidder chooses to pay for an acquisition, whereas the source of financing is what
the bidder funds an acquisition. The means of the bidder uses to get the funds can be different
from their payment decisions in M&As. Whilst conventional M&A studies considered the
means of payment as the same as the source of financing, Martynova and Renneboog (2009)
disentangled the effect of the source of takeover financing and the effect of the method of
payment. According to their study, the source of financing includes purely internal funds,
equity issues, debt issues and combinations of debt and equity issues. Acquisitions are paid
with cash when the source of financing is purely internal funds and debt issues, and are paid
with stocks when the sourcing of financing is equity issues, or a combination of equity and
debt and cash. Analysing a sample of 1,361 completed acquisitions in 26 European countries15,
the authors found that acquisitions funded by internal sources underperformed those funded by
debt issues.
15 Their sample consists of both European domestic and intra-European cross-border acquisitions with both
bidding firm and target firm from European countries or the U.K.
Chapter 2. Literature Review: M&A Process
50
Pecking order theory is commonly violated in the context of M&As especially when bidders
are financially constrained or distressed. Martin (1996) noted that one of the main reasons that
a firm uses stocks as a means of payment for acquisitions is that the firm is in short of cash
prior to the M&A announcement date. Jensen (1986) found supportive evidence that a large
free cash flow motivates managers to use cash instead of stocks to finance an acquisition.
Harford et al. (2009) indicate that firms that have a higher leverage ratio relative to targets tend
to use equity as a means of payment for finance acquisitions while the lower leverage ratio
firms tend to use debt-financed cash. Their study relates the method of payment to the capital
structure of the firm. Firms in the face of larger risks of financial distress or bankruptcy are
more cautionary about the change in their leverage, indicating an important role of risks played
in the method of payment choices. In a similar vein, Alshwer et al. (2011) found financially
constrained bidders tend to retain internal sources to maintain financial flexibility.
Other studies have related the method of payment choices to managerial control. Managers
holding large stocks would have extensive control over the firm, in that they should be reluctant
to offer stocks in order to maintain their control. Stulz (1988) suggests that managers attemping
to protect their ownership of the firm avoid using stocks as a means of payment for acquisitions.
Martin (1996) obtained results consistent with of those in Stulz’s work (1988). The results
suggest a non-linear relationship between managerial ownership and the method of payment.
Managers are insensitive about the method of payment when their ownership is low, while a
high level of managerial ownership, between 5 and 15 percent, reduces the probability of
acquisitions being paid for with stocks, as managers with large stock holdings avoid using
stocks as this result in dilution of their control to over the firm. Similarly, Faccio and Masulis
(2005), who investigated a sample of European M&As between 1997 and 2000, noted that cash
is more likely to be used to fund acquisitions when the level of managerial ownership is
Chapter 2. Literature Review: M&A Process
51
between 20 and 60 percent. These authors also document that the probability of stock payments
decreases when target firms have concentrated managerial ownership, since this creates new
block holders threatening the firm control of bidder managers.
However, it is argued that neither the target nor the bidder managerial ownerships have an
impact on the method of payment. Zhang (2001) examined the effect of managerial ownership
on the method of payment in the U.K. market using a number of factors that have a significant
impact on the method of payment, including the firm size, the financial policy of the bidder,
returns of equity and the performance of the bidder.16 Taking these controls into account, the
ownerships of the two firms involved showed little impact resulting from the method of
payment. The author concluded that there was little evidence of the managerial ownership
effect on the method of payment, but he failed to explain the rationale behind this finding.
2.2.2.2.4. Behavioural finance
Studies have emphasised the role of managerial behaviour played in the method of payment
choices. Shleifer and Vishny (2003) suggest that bidder managers choose an appropriate means
of payment for acquisitions. By doing so, bidder managers utilise stocks in acquisitions when
their firms are overvalued. Additionally, Vijh and Yang (2013) found small public targets were
less vulnerable to overvalued stocks. Studying a large sample of U.S. public acquisitions
between 1981 and 2004, the authors established a positive relationship between the target size
and stock-financed acquisitions, suggesting that small targets leave bidders limited room to
dilute overvaluation and acquiring small targets is associated with larger transaction costs and
higher offer premiums than large targets. This suggests that the valuation of firms is an
16 Zhang (2001) found that stocks were likely to be used for acquisitions when bidder firms were large and equity
returns were higher (as bidding firms might believe their stocks are undervalued therefore paying for acquisitions
with cash), while cash was likely to be employed for acquisitions when bidder firms paid higher dividends to their
shareholders and the bidder firm performed better in the stock market.
Chapter 2. Literature Review: M&A Process
52
important factor for managerial considerations regarding the method of payment choices. Once
again, their evidence suggests rational managers link the method of payment to the valuation
of the firm.
Rhodes-Kropf and Visvanathan (2004) also claim misvaluation drives all stock-financed
acquisitions. Rhodes-Kropf et al. (2005) identified the three sources of valuation errors that
drive stock-financed acquisitions. Of these, the firm- and sector-level valuation errors were the
main explanatory factors. In support of this, Dong et al. (2006) found that stocks were likely
to be used for acquisitions as the two firms involved had a high valuation. They measured the
firms’ valuations with P/B and P/V, which are based on managerial perceptions, reasoning that
‘It could be argued that it is not actual misvaluation that influences takeovers, but merely an
incorrect perception by managers that there is overvaluation. If managers believe that the
price-to-book ratio is an indicator of misvaluation, they may make takeover decisions based
on this measure.’
Dong et al. (2006:756)
Following this, Ben-David et al. (2015) used short interest as a measure for mispricing where
short interest is a sophisticated investors’ belief regarding a firm’s misvaluation. The authors
suggest that bidders engaging in stock-financed acquisitions have stronger short interest, which
distinguishes the measure used for the real investment opportunities of the firm. The
mechanism is that short-sellers tend to take (avoid) a short position in overvalued (undervalued)
stocks. Further, their study suggested that firms in the top quintile of short interest are 54
percent more likely to engage in stock-financed acquisitions, which is consistent with the
misvaluation hypothesis.
Chapter 2. Literature Review: M&A Process
53
Studying the willingness of targets to accept overvalued stocks, Burch et al. (2012) found that
institutions tend to accept or hold stocks with high valuation. It was found more than 56 percent
of the institutions in their sample sell target stocks before the completion of an M&A or convert
them to bidder stocks soon after that. Contradicting the view that targets accept overvalued
stocks by mistake (Rhodes-Kropf and Visvanathan, 2004)17, Burch et al. (2012) found that
institutional shareholders time the market using highly valued stocks. Managers find it easier
to use overvalued stocks to liquidate their holding position. They can sell stocks before the
market downgrades the value of stocks. In addition, they presumably believe that highly valued
firms are also well-performing firms or those with better investment opportunities. Hence,
holding stocks in these firms can generate benefits in the future.
Studies that examine the misvaluation hypothesis largely borrow MTBV as a measure for
mispricing, (Rhodes-Kropf and Visvanathan, 2005; Dong et al., 2006). However, Di Giuli
(2013) argued that MTBV is problematic for misvaluation as it is a proxy for both mispricing
and investment opportunities. Specifically, a firm that has better investment opportunity or is
highly valued leads to the same prediction that its MTBV is high. This is even misleading in
industries of high growth prospects. So as to disentangle the two effects, the author proposed
a new measure for investment opportunities, which is the average ratio of capital expenditures
over assets in the four years following the merger. The proxy is based on post-merger
investments. Using this on a sample of 1,187 mergers between 1990 and 2005, the author found
that one standard deviation increase in capital expenditure increases the probability of paying
acquisitions with stocks rather than cash by 7.5 percent, suggesting that better investment
opportunities lead bidders to increase stock payments for acquisitions.
17 The mistake refers to an overestimation of the synergy or underestimation of the market-wide overvaluation.
Chapter 2. Literature Review: M&A Process
54
Rather than the misvaluation hypothesis, which predicts cash as a means of payment for
undervalued targets, the supply-driven explanation proposed by Zhang (2009) suggests that
cash is likely to be used as a means of payment for acquisitions when the popularity of cash
for target shareholders is high and target firms are less active and more constrained. Offering
what the target shareholders want, bidders also reinforce their bargaining position in an M&A
deal. It was found that cash paid to target shareholders with greater cash popularity is associated
with lower offer premiums.
Additional factors also exhibit a significant impact on the method of payment include. First,
the size of the firm relates to the method of payment. Dong et al. (2006) suggest that equity is
more likely to be used for acquisitions involving large bidder and target firms. It can be
interpreted as follows: bidders may find it hard to borrow large amount of cash to finance a
large target and thus use stocks as a substitute (Martin, 1996; Moeller et al. 2004). Further,
tender offers are more likely to be associated with cash payment whereas mergers are more
likely to be associated with stock payments (Martin, 1996). In addition, cash is likely to finance
hostile takeovers and generate higher abnormal returns for bidders (Schwert, 2000). In a cross-
border M&A settlement, the legal environment differences across the countries also affect the
method of payment (Martynova and Renneboog, 2009).
Chapter 2. Literature Review: M&A Performance
55
2.3. M&A performance
M&A performance is the consequence of M&As, which is one of the direct criteria that assess
the success of an M&A deal. It is a well-documented fact that targets receive positive market
reactions around the M&A announcement date, whereas bidders on average do not gain (Dodd
and Ruback, 1977; Jensen and Ruback, 1983; Rau and Vermaelen, 1998; Fuller et al., 2002;
Moeller et al., 2004; Alexandridis et al., 2013). As abnormal returns for short-term period may
not fully capture the announcement effect of a takeover, studies have also assessed the long-
term performance of the firm (Loughran and Vijh, 1997; Rau and Vermaelen, 1998; Agrawal
and Jaffe, 2000), since it reflects the holding experience of the investors. Some studies have
also assessed the operating performance of the firm in the post-merger period, which is an
accounting-based approach (Healy et al., 1992; Ghosh, 2001; Andrade et al., 2001; Martynova
et al., 2006). Overall, M&A performance is sensitive to the sample period examined18, the
sample criteria selected and the benchmark models employed to calculate abnormal returns of
a firm.
A number of factors have had a significant impact on the abnormal returns of the firm, and
been summarised into two broad categories: firm-specific characteristics and deal-specific
characteristics. Studies that link the M&A performance of firms include those variables as
standard controls, such as MTBV (Rau and Vermaelen, 1998; Sudarsanam and Mahate, 2003),
the size of the firm (Asquith et al., 1983; Moeller et al., 2004; Alexandridis et al., 2013), the
method of payment (Travos, 1987; Franks et al., 1988; Shleifer and Vishny, 2003), target status
(Fuller et al., 2002; Antoniou et al., 2007), merger choices (Dodd and Ruback, 1977;
18 There are six merger waves defined in the M&A literature (Martynova and Renneboog, 2008). The first merger
wave was between 1890 and 1903, the second merger wave was between 1910 and 1929, the third merger wave
was between 1950 and 1973, the fourth wave is between 1981 and 1989, the fifth wave was between 1993 and
2001, with the sixth merger starting in 2003.
Chapter 2. Literature Review: M&A Performance
56
Berkovitch and Khanna, 1991), deal attitude (Schwert, 2000; Martynova and Renneboog, 2006)
and industry relatedness (Morck et al., 1990; Agrawal et al., 1992).
2.3.1. Short-term performance
Numerous studies have examined the abnormal returns for the bidder firm, target firm, and the
combined firm with the event study. This approach has become dominated in analysing the
abnormal returns of the firms since it was first adopted by Fama et al. (1969). According to
Jensen and Ruback (1983), abnormal returns are the difference between the realised returns of
the firm and the benchmark returns:
‘Abnormal returns are measured by the difference between actual and expected stock returns.
The expected stock return is measured conditional on a realised return on a market index to
take account of the influence of market-wide events on the returns of individual securities.’
Jensen and Ruback (1983:6)
Early event studies measure abnormal returns of the firm around the effective date of merger.
However, Dodd and Ruback (1977) argued that the price effect could be released prior to the
announcement date, leading the price effect surrounding the completion date to be insignificant
(Asquith, 1983). Martynova and Renneboog (2011) proposed the necessity of observing both
the pre-announcement and post-announcement returns in an M&A study, since both are able
to measure M&A performance of the firm. Specifically, pre-announcement returns could
capture the information leakage, insider trading and market anticipation of an M&A deal while
post-announcement returns could reflect market correction for investor sentiment, such that
they over- or under-react to the M&A deal.
Chapter 2. Literature Review: M&A Performance
57
Martynova and Renneboog (2011) hold that M&A information is partially released prior to the
M&A announcement date, suggesting that investors could learn about the bidder’s objective of
takeover, the target’s attitude towards to the bid and so forth prior to the announcement date.
As such, abnormal returns surrounding the announcement date of a merger may not be able to
fully reflect the market reactions to an M&A bid. Following this view, Schwert (1996)
measured investors’ reaction to the M&A announcement with stock price change in a period
of 42 trading days prior to the announcement date and the announcement date. Stock price
change in the post-announcement period captures additional information that pre-
announcement period fails to do, such as, the level of post-merger integration and the
performance of successful and failed deals. However, it is difficult to assess the post-
announcement effect separate from other effects that also lead to a change of stock price
following a merger, such as financial or operational performance (Martynova and Renneboog,
2008). In spite of these factors, studies that examine the wealth creation from M&As have
employed a relatively short event window holding that price change around the announcement
date fully reflects M&A performance (Asquith, 1983; Dodd, 1980).
M&A studies have revealed appealing features of abnormal returns of firms by merger waves.
It is well-established that M&As generally create value for the two firms involved prior to the
fifth merger wave starting in 1993. Investigating the price change of the entire merger process
between 1962 and 1976, Asquith (1983) found positive announcement returns for bidder and
target firms regardless of deal outcomes. More specifically, targets enjoyed 6.2 percent two-
day excess returns19 for successful deals and this figure up to 7 percent for unsuccessful deals
19 The two-day excess return is the sum of abnormal returns in the window one day prior to the takeover
announcement date and the takeover announcement date.
Chapter 2. Literature Review: M&A Performance
58
while bidders enjoyed 0.2 percent two-day excess returns for successful deals and 0.5 percent
for unsuccessful deals.
Reviewing thirteen studies examining abnormal returns of both bidders and targets between
1950s and 1980s, Jensen and Ruback (1983) found that targets gain from M&As, whereas
bidders do not loss on average in abnormal returns around the announcement date. Of particular
note, targets outperform bidders in the samples of successful and unsuccessful M&A deals,
whereas bidders generally gain in successful M&A deals. In addition, bidder abnormal returns
are high regardless of whether the merger choice is a tender offer or merger for successful
M&A deals, and the two firms suffer small losses in abnormal returns for failed tender offers
and mergers. Franks and Harris (1989), who conducted an out-of-sample test, reported target
announcement returns in the range between 25 and 30 percent while bidders earn zero or
modest gains based on a sample of 1,800 U.K. takeovers announced in the period of 1955 to
1985, consistent with those in the work of Jensen and Ruback (1983).
Studies have shown significantly positive bidder announcement returns prior to 1993.
Analysing a sample of U.S. tender offers between 1958 and 1978, Dodd and Ruback (1977)
found that bidders earned 2.83 percent abnormal returns in a period of the announcement date
until 20 days following a takeover. Focusing on a similar sample period of Dodd and Ruback
(1977), Franks et al. (1977) reported abnormal returns of 4.6 percent for U.K. bidders. Studying
a comprehensive U.S. M&A sample, Schwert (1996) found both the two firms earn
significantly positive announcement returns regardless of merger choice. Eckbo and Thorburn
(2000), focusing on a sample consisting of both domestic and cross-border M&As between
1964 and 1983, found that abnormal returns to a U.S. bidder’s acquisition of a Canadian firm
were zero, whereas a Canadian bidder earned significantly positive in domestic M&As. Jensen
Chapter 2. Literature Review: M&A Performance
59
(1988) explained positive bidder abnormal returns as the effectiveness of the market for
corporate control returns prior to 1990s when managers compete to improve the efficiency of
the firm through M&As.
Whilst studies have found significantly negative bidder abnormal returns around the
announcement date after the start of the fifth merger wave, identifying various sources of value
destruction, such as agency problems, managerial overconfidence and market mispricing.
Agency problems suggest that the value of the firm is likely to be transferred from a firm that
is associated with severe conflicts of interests between the shareholder and the manager.
Harford et al. (2012) found entrenched bidder managers engaging in a value-destroying M&A
for fear of their stocks being diluted avoid using all equity, which in return destroys the firm’s
value. Moreover, entrenched managers avoid biding private targets that can potentially create
value for the combined firm, and they overpay for public firms and firms of low synergies.
Doukas and Petmezas (2007), who examined the effect of managerial overconfidence on bidder
gains between 1980 and 2004, found that bidder announcement returns were lower for a higher
order acquisition than a lower order acquisition, explaining this as the managerial self-
attribution bias. Using a large sample of U.S. mergers between 1978 and 2000, Dong et al.
(2006) found negative abnormal returns for overvalued bidders’ acquisition of a less
overvalued target. In such a case, bidder managers sacrifice the wealth of shareholders in the
short term to protect the wealth of the shareholders in the long term, and this finding is also
supported by Ang and Cheng (2006).
The sixth merger wave that started in 2003 witnessed some similar patterns of M&A
performance as those in the fifth merger wave. Alexandridis et al. (2010) suggest that bidder
abnormal returns decrease with the competitiveness of the takeover market. Studying a
Chapter 2. Literature Review: M&A Performance
60
worldwide M&A sample consisting of M&As announced between 1990 and 2007, the authors
suggest that bidders in the takeover competitive markets (i.e. U.S., U.K. and Canada) tend to
overpay, resulting in loss in the announcement returns, whereas bidders beyond these three
countries on average realise gains. Their study suggests that bidder abnormal returns vary
significantly from market to market. In spite of lower offer premiums are paid for targets during
the sixth merger wave than the fifth merger wave, bidders do not gain. Alexandridis et al. (2012)
underlined this to free cash flow problem, suggesting that the firms undertaking M&As in the
sixth merger wave tend to have lower valuation and higher cash balances than those in the fifth
merger wave, indicating that cash payment dominates stock payments in this period.
2.3.2. Long-term performance
2.3.2.1. Stock performance
A large body of M&A research has documented significant negative bidder abnormal returns
in the long-term up to a five-year period following a merger (Asquith, 1983; Jensen and Rubuck,
1983; Agrawal et al., 1992; Loughran and Vijh, 1997; Rau and Vermaelen, 1998; Franks and
Harris, 1989) 20. Asquith (1983), examining the bidder’s performance in the post-acquisition
up to 240 trading days following a takeover announcement date, reported a dramatic decline in
bidder abnormal returns compared with those on the outcome date of a merger. Specifically,
bidder abnormal returns decreased by 1.9 percent in the 100 days and by 7.2 percent in the 240
days following the date of the deal outcome. Jensen and Ruback (1983) reported negative
bidder abnormal returns of -5.5 percent in the year following a merger and noted that the result
is unsettling as it indicates market inefficiency. They further suggested this finding as the
market’s overestimation of the synergies accompanying a merger. Ruback (1988) late accepted
20 Despite these studies, Higson and Elliott (1998) reported zero abnormal returns in the three years following a
merger to the U.K. bidders between 1975 and 1990. In addition, equal-weighted abnormal returns to bidders were
found to be insignificant while bidders whose abnormal returns were calculated with the value-weighted returns
were positive and small.
Chapter 2. Literature Review: M&A Performance
61
the result that significant negative abnormal returns to bidders over the two years following a
merger as a fact. The notion that the market is inefficient has been supported by many
additional studies in favour of negative bidder long-term abnormal returns. Agrawal et al.
(1992), among other researchers, also reported negative bidder abnormal returns of -10 percent
over five years after a merger and their results continue to hold after controlling for size effect
and beta risk in various benchmark models.
Agrawal and Jaffe (2000), citing the work of Agrawal et al. (1992), attribute underperformance
as a ‘post-merger performance puzzle’. The authors outlined the method of payment and the
performance extrapolation hypotheses to this underperformance puzzle. The former hypothesis
indicates that when equity is used, the bidder is likely to be overvalued, and when cash is used,
the bidder is undervalued. Since the overvaluation will be corrected by the market in the long
term, stock-financed acquisitions are associated with negative market reactions, while cash-
financed acquisitions generate positive market reactions. The performance extrapolation
hypothesis indicates that the market overestimates the firms with high MTBV, which creates a
strong momentum around the takeover announcement date and leads to a strong long-term
reversal. Thus, Agrawal and Jaffe’s findings (2000) are consistent with those of Rau and
Vermaelen (1998). In a similar vein, Gregory (1997), who examined the relationship between
long-term abnormal returns and the free cash flow of a firm in the U.K. market, noted that firms
with high Q and low free cash flow are overvalued and underperform in the market in the long
term.
Another explanation for bidders’ underperformance in the long term is the neglect of the post-
merger integration process. Hansen (1987) indicates larger potential revaluation loss is
associated with the greater relative size of the target to the bidder. Moeller et al. (2004) suggest
Chapter 2. Literature Review: M&A Performance
62
that large deals require more diligent work by bidders, and they are associated with
considerable integration costs that lead to downward synergies. Despite their study not directly
addressing the long-term performance of the bidders, managers of a larger firm are likely to be
hubris-infected due to extensive managerial autonomy in the firm, causing more negative
announcement returns. Likewise, Antoniou et al. (2008) re-addressed the loss in the long-term
abnormal returns as the neglect of post-merger integration in a sample of U.K. public
acquisitions.
Others attribute bidders’ underperformance in the post-merger period to misspecifications of
the benchmark models employed. Franks et al. (1991), who tested a sample of 399 U.S.
takeovers between 1975 and 1984, suggested underperformance is measurement errors of the
models. The authors use multiple benchmark models in assessing the long-term performance
of the bidder and find insignificant abnormal returns, and the results are significantly different
depending on which portfolio is used: the equally- or value-weighted. Moreover, the authors
claim that long-term abnormal returns are largely dependent upon the relative size of the target
and the bidder, the method of payment, and the models employed. Barber and Lyon (1997)
also addressed bad model problems. They proposed multifactor models instead of a single
model to avoid mean-variance inefficiencies, and control firms of similar size and MTBV to
calculate long-term abnormal returns within five years following a merger. Their approaches
showed well-specified test statistics.
2.3.2.2. Operating performance
Researchers also measure bidder long-term performance using operating performance by
assessing the firm’s performance from its fundamentals, avoiding measurement errors from the
benchmark models based on an efficiency market. (Healy et al., 1992; Barber and Lyon, 1996;
Chapter 2. Literature Review: M&A Performance
63
Ghosh, 2001; Andrade et al., 2001; Martynova et al., 2006). According to Powell and Stark
(2005), previous M&A studies for assessing a firm’s operating performance following a merger
were prepared from three perspectives: the measures, the deflator choices21 and benchmark
models. Barber and Lyon (1996) suggest the important roles that firm size and pre-
announcement performance played in operating performance, showing test statistics well when
controlling for these two factors in the matched firm sample.
Studies commonly employ cash flows as a proxy to assess a firm’s operating performance
following a merger (Healy et al., 1992; Barber and Lyon, 1996), as other accounting-based
measures are likely to be manipulated after takeovers (Powell and Stark, 2005). Healy et al.
(1992), examining a sample of 50 largest U.S. publicly industrial mergers completed between
1979 and mid-1984 and using operating cash flows and asset productivity as proxies for
operating performance, found strong operating improvements following a merger. In a further
analysis, the authors found a positive relationship between abnormal returns and operating cash
flow performance of the bidder, implying that the market can predict the operating performance
of the firm.
Likewise, Switzer (1996) in his study focusing on a relative large sample size finds firms’
operating performance improved following a merger, regardless of the firm’s size, relatedness
of the two firms and bidder leverage. Linn and Switzer (2001) also provide evidence in support
of the existence of improvement in a post-merger performance. According to their study, cash
payment is associated with significantly greater change in operating performance than stock
payments. Their results indicate that bidders use cash as they have private information about
21 Powell and Stark (2005) have reviewed a series of studies relating to the techniques used in scaling the
accounting measures.
Chapter 2. Literature Review: M&A Performance
64
the potential synergies to be realised in the long term, reflected in an improvement of operating
performance.
However, Ghosh (2001) challenges the findings of Healy et al. (1992) by arguing that such an
improvement in operating performance is due to the fact that firms engaging in acquisitions are
larger firms compared with their industry peers. In most cases, large firms enjoy superior
performance, thus examining the operating performance of the firm exists bias when including
large firms. To combat this concern, the author uses firms matched on size and performance as
a benchmark to account for pre-performance of the bidder and compares the operating
performance of the firm before and after mergers with the matched firm sample. His study
shows little evidence that operating cash flow performance improves following a merger.
Many additional studies borrowing different operating measures have shown a decline to
operating performance following a merger (Mueller, 1980; Clark and Ofek, 1994; Fu et al.,
201322). Mueller (1980) observed a decrease of operating performance measured with ROE,
sales growth rate and total asset growth rate. Clark and Ofek (1994), studying a sample of
M&As with financially distressed targets and using earnings before interests, taxes and
depreciation (EBITD) as a proxy23 for operating performance also document a decline in
operating performance.
Using different accounting-based measures for the firm’s operating performance, Gugler et al.
(2003) reported mixed results in a sample of worldwide M&As announced between 1981 and
1998. More specifically, measuring operating performance with Profit/Assets, the authors were
22 Fu et al. (2013) detected the operating performance by focusing on a test for an implication of the misvaluation
hypothesis of Shleifer and Vishny (2003), which is whether the long-term performance of the stock bidders
improves when they use overvalued stocks as a means of payment for acquisitions. 23 His study scaled EBITD with revenues, which is EBITD/Revenues.
Chapter 2. Literature Review: M&A Performance
65
able to show an improvement in operating performance while the results decreased when
measured with Sales/Assets. Similarly, Martynova and Renneboog (2006) used four different
accounting-based measures to detect the operating performance for a sample of European and
U.K. M&As between 1997 and 2001. Controlling for industry, size and pre-takeover
performance in the matched firm sample, the authors showed that the previous significant
decrease in operating performance tends to be insignificant. In addition, the authors indicate
when the intercept model (or the regression-based model) was applied, the profitability tended
to be significantly positive and when the change model used the profitability tends to decrease.
Once again, their studies indicated that a firm’s long-term operating performance was sensitive
to the factors controlled and benchmark models employed.
2.3.3. Factors influencing M&A performance
2.3.3.1. The method of payment
The performance of the firms involved in an M&A deal are dependent upon a number of factors
examined in M&A literature (Datta et al., 1992; King et al., 2004; Eckbo, 2009), and the
method of payment has been widely studied in this regard. A review of the method of payment
hypotheses was presented in the 2.2.4 section of this thesis.
There is unanimous agreement among researchers that cash payment outperforms stock
payments in the short term. Studying a sample of U.S. acquisitions in the period of 1980 to
2001, Moeller et al. (2004) reported 1.38 percent bidder abnormal returns for acquisitions were
financed by cash, which is higher than stocks (0.15%). Martynova and Renneboog (2006), who
employed an 11-day window to calculate the bidder cumulative abnormal returns in a sample
of E.U. acquisitions, also reported quite similar results. In their study, though stock-financed
Chapter 2. Literature Review: M&A Performance
66
acquisitions can generate positive abnormal returns to bidders, they are significantly lower than
the returns on cash payment, 11.1 percent as opposed to 20.17 percent.
Similar results were reported in the M&A studies that focus on the bidders’ long-term abnormal
returns. Loughran and Vijh (1997) examining 947 acquisitions in the period 1970 to 1989 noted
that stock-financed mergers significantly underperformed cash-financed tender offers in the
five years following a merger, -25 percent as opposed to 65 percent. These authors provided
two proposed reasons in support of this finding. First, it was found that M&As that are paid in
cash are also tender offers that replace inefficient management and thus improve a firm’s value.
Secondly, bidders pay with stocks if the stocks are overvalued, leading to a decline in value
after market correction. In this vein, Bouwman et al. (2003), examining a sample of U.S.
acquisitions between 1979 and 1998 and using a two-year period following the acquisition,
reported negative excess returns of 7.03 percent for all stock-financed mergers whereas there
were insignificant excess returns of -1.76 percent for their cash counterparts. Analysing bidders’
long-term abnormal returns in a three-year period following an acquisition in a U.S. sample
between 1984 and 2001, Ang and Cheng (2006) reported negative excess returns of -12.45
percent for all-stock financed acquisitions, the result the authors explained with the
misvaluation hypothesis. Sudarsanam and Mahate (2003), focusing the U.K. market, reported
significantly negative abnormal returns for stock-financed acquisitions in the period 2 to 36
months following an acquisition regardless of the level of the price to bidders’ earnings ratio.
Researchers have addressed the effect of the payment method on the firm’s performance by
different M&A subsamples. Bouwman et al. (2003), who examined bidder abnormal returns
by merger choice, indicated that abnormal returns to bidders were lower for acquisitions
financed by stocks compared with those financed by cash, the result is more significant for the
Chapter 2. Literature Review: M&A Performance
67
merger subsample than the tender offer subsample. In a sample of U.S. acquisitions between
1979 and 1998 and to calculate bidder abnormal returns with a 3-day window, the authors
demonstrated insignificant positive abnormal returns for a sample of tender offers financed by
cash and stocks: 0.36 percent and -0.62 percent, while for those mergers, bidders enjoy
significantly positive abnormal returns for cash-financed acquisitions (0.88%) and significantly
negative abnormal returns for stock-financed acquisitions (-0.79%).
Whilst numerous M&A studies documenting evidence show that cash payment generates
positive abnormal returns and stock payments generate negative abnormal returns for bidders,
Chang (1998) reported significantly negative bidder abnormal returns for stock-financed public
acquisitions (-2.46%), but insignificantly negative bidder abnormal returns for cash-financed
public acquisitions (-0.02%). However, these results are reversed when targets are private firms
(i.e. unlisted firms). In particular, bidder abnormal returns for stock-financed private
acquisitions are significantly positive (2.64%) whereas cash-financed private acquisitions are
insignificantly positive (0.09%). This author provided three hypotheses to underline this
finding: the limited competition hypothesis, the monitoring hypothesis and the information
hypothesis. The limited competition hypothesis indicates that positive bidder abnormal returns
are due to the high probability of underpayment given weak competition for private
acquisitions. The monitoring hypothesis indicates outstanding performance achieved by
bidders using stocks for private acquisitions, as paying with stocks creates new block holders
that motivate the firm’s performance. The information asymmetry hypothesis indicates that the
market may re-evaluate the bidder’s valuation and reacts positively to the bid announcement
based on the rationale that private targets value stocks with caution. This finding is also
supported by Draper and Paudyal (2006) who have examined the U.K. takeover market where
a large majority of deals involve a private target firm.
Chapter 2. Literature Review: M&A Performance
68
2.3.3.2. Target status
Target status plays an important role in explaining bidder announcement returns. In most cases,
a private target is an illiquid firm, motivating the bidder to restore price discounts through
takeovers. Furthermore, a private target has a weak bargaining position, which reduces M&A
offer premiums. Studying a sample of 3,135 acquisitions between 1990 and 2000 and using a
5-day window for CARs, Fuller et al. (2002) found significantly positive returns were created
for bidder firms. When looking into subsamples according to target status, the authors noted
that bidder announcement returns were -1.0 percent when the target was a publicly-listed firm,
2.1 percent when the target was a privately-held firm and 2.8 percent when the target was a
subsidiary. Results are similar to Moeller et al.’s work (2007) employing a 3-day window for
CARs. Moller et al. (2007) studied a sample of 4,322 U.S. acquisitions announced between
1980 and 2002 and found that for all stock-financed acquisitions, the value of the bidder
declined by 2.3 percent when the target was a public firm, whereas it increased by 3.4 percent
when the target was a private firm. Their results are also supported by Conn et al. (2005) who
reported significantly loss in bidder abnormal returns in acquisitions involving a public target
when analysing a sample of U.K. M&As between 1984 and 2000.
Contradicting this line of studies, Bradley and Sundaram (2004) show that private firms
decrease the wealth effect of the bidder. Analysing a sample of over 12,000 U.S. acquisitions
between 1990 and 2000 and measuring abnormal returns in a window of 1 day and 24 days
after the M&A announcement date, the authors found that bidder announcement returns were
significantly negative (-10.09%). In particular, all-stock payments decreased the value of the
bidder firm by 6.35 percent for public acquisitions, whereas a decrease of 14 percent was noted
for private acquisitions. It can be argued that private firms are associated with high risks as
Chapter 2. Literature Review: M&A Performance
69
their information is less transparent than public firms, hampering synergy estimation of the
target firm.
2.3.3.3. Merger relatedness
Unrelated acquisitions24 refer to the two firms involved in an M&A deal being from different
industries. Researchers have not reached a consensus on whether diversifying acquisitions
improves the value of a firm. Datta et al. (1992) in a review study identified the wealth of
shareholders in a related acquisition was created through a transfer of core skills, economies of
scale, economies of scope and the value enhancement from the market power. The wealth of
shareholders in an unrelated acquisition is created through risk diversification, cheaper access
to capital, and value enhancement resulting from operational efficiency.
Morck et al. (1990) found diversified acquisitions are associated with lower bidder abnormal
returns, reasoning that there was a lack of experience when managers step into an unrelated
field. In support of this, Berger and Ofek (1995), by comparing the stand-alone value of the
firm and the market value of the firm in a diversified M&A deal, find that the market value of
the firm following a diversified acquisition was on average between 13 percent and 15 percent
loss in abnormal returns. Denis et al. (2002) claimed that neither global diversification strategy
nor industrial diversification strategy can create value for the firm. Contradicting this thread of
literature, Comment and Jarrell (1995) noted that corporate diversification increased the wealth
of shareholders in 1980s, identifying that diversification strategy as enhancing the firm’s value
through economies of scale.
24 Also known as diversified or conglomerate acquisitions.
Chapter 2. Literature Review: M&A Performance
70
2.3.3.4. Merger attitudes
The attitude of a merger can be broadly split into two, namely, solicited (friendly) and
unsolicited (hostile). A hostile deal refers to a deal opposed by the board of the target
management. Servaes (1991) reports loss in bidder announcement returns of hostile deals,
reasoning that the hostile bid results in high offer premiums or an increase of the probability
that a deal will fail. And in either case, bidders receive negative market reactions. In contrast,
friendly bids create value for both the bidder and the target firms, according to Morck et al.
(1988).
Schwert (2000) wrote that “the distinction between hostile and friendly offers is largely a
reflection of negotiation strategy” and “hostile takeovers mean different things to different
people”, implying that hostile takeovers do not reveal useful information to the market. The
author finds little evidence of the relationship between bidder announcement returns and hostile
deals based on the SDC data. The findings support the view that target managers opposing
hostile bids either because they are entrenched or because they use it as a bargaining strategy,
which contradicts Lang et al.’s view (1989) that target managers opposing hostile takeovers
signals to the market that their firm is well managed and creates value for the combined firm.
2.3.3.5. Merger choice
The choice of an acquisition includes tender offers and mergers. Tender offers are a share
purchase directly proposed to target shareholders, whereas mergers involve a negotiation
between the boards of the two firms involved. There are two advantages to initiating an
acquisition with tender offers, rather than mergers. First, tender offers avoid lengthy
negotiation processes. Golubov et al. (2012) indicate that the speed of deal completion for
tender offer is greater than that of the merger. They found that tender offers are approximately
Chapter 2. Literature Review: M&A Performance
71
55 days shorter than mergers in a sample of completed bids: 1,650 completed deals out of a
1,836 M&A sample. Secondly, initiating tender offers does not need the prior approval or a
contract with the target management (Betton et al., 2008), reducing bidder managers’ efforts.
However, tender offers are relatively costly for bidders. Offenberg and Pirinsky (2015), by
analysing a sample of hand-collected U.S. acquisitions announced between 2007 and 2012,
found that tender offers were associated with higher offer premiums than mergers since tender
offers may indicate the market that targets would generate large synergies for the combined
firm. In this sense, the authors should expect a positive relationship between tender offers and
bidder gains. However, they do not provide direct evidence in support of this implication.
Instead, they find that tender offers incur negative bidder abnormal returns, as bidders initiating
a tender offer are regarded as bad news. Moeller et al. (2004) support their view with positive
bidder abnormal returns associated with tender in a sample of completed U.S. acquisitions25.
The finding is more appealing for tender offers involving a larger bidder than a smaller bidder
(Offenberg and Pirinsky, 2015). It can be concluded that tender offers, though costly, reduce
the speed of the M&A process that turns out to be effective for M&As.
It is expected the use of tender offers increase the firms’ M&A performance. Jensen and
Ruback (1983) indicate that tender offers are associated with higher bidder and target
announcement returns than those of mergers. They reported 29 percent of abnormal returns to
target shareholders in tender offers compared with 16 percent to mergers, and 4 percent in
tender offers compared with zero gains in mergers. Franks and Harris (1989) reported 24
percent abnormal returns in the month of the announcement date to the target shareholders,
25 Offenberg and Pirinsky (2015) suggest that tender offers are more likely to be associated the deal is completed
than mergers.
Chapter 2. Literature Review: M&A Performance
72
which are significantly higher than those recorded in tender offers (14.8%). Bidders also enjoy
high abnormal returns in tender offers, reporting a positive 1.2 percent abnormal returns in
tender offers, higher than those in other offers (-3.6%).
Many additional studies have provided explanations of the fact that abnormal returns in tender
offers are higher than those of mergers. Bradley et al. (1988) suggest positive target returns in
tender offers are a result of increased competitiveness of the market, reasoning that the
introduction of the William Act in 1968 is the main driving force. However, this view has been
challenged by Franks and Harris (1989), who found positive returns to the U.K. targets both
before and after 1968. Martin (1996) and Dong et al. (2006) suggest that tender offers are likely
to be those acquisitions financed by cash that conveys the market good news. Kohers et al.
(2007) suggest bidders’ propensity for using tender offers is larger in the cases where targets
are defensive and there are multiple bidders. This line of studies suggests that bidders can better
explore synergies when proposing tender offers.
2.3.3.6. Size and relative size
The size of each firm involved in an M&A deal is also an important factor influencing M&A
performance of the firm. Agrawal et al. (1992) were the first to account for the size of the firm
when examining the bidder long-term abnormal returns, as there are a large number of large
firms in their sample. Further to this, Moeller et al. (2004) find bidder size is a key factor in
explaining bidder’s performance. These authors examined the relationship between bidder size
and firm performance in the M&As and find that managers of large bidders tend to destroy
more of the firm’s value than their smaller counterparts, reasoning that managers of large firms
who do not have any restrictions regarding using the firm’s resources potentially dissipate these
Chapter 2. Literature Review: M&A Performance
73
resources in M&As. Evidence that bidder size and bidder gains are negatively correlated is also
supported in prior M&A studies (Eckbo and Thorburn, 2000; Moeller et al., 2005).
In addition, target size is also a crucial factor in assessing M&A performance. Analysing a
sample of 3,691 U.S. mergers between 1990 and 2007, Alexandridis et al. (2013) find a
negative relationship between target size and bidder gains. In particular, announcement returns
to large targets were 2.37 percent lower than small targets and bidder announcement returns
decrease by 1.1 percent for every standard deviation increases in target size. The authors
explain this as post-merger integration problem of large deals, as they are complex, hence, they
are hard to create value for the firm. According this theme in the literature, relative size as well
as the size of the two firms are essential in examining M&A performance of the firm and should
therefore be included (Agrawal et al., 1992; Loderer and Martin, 1992).26
Relative size is the deal value scaled by the bidder value, which captures the materiality of the
deal decision to the bidder. A larger ratio indicates that the M&A decision is relatively more
important to the bidder. Relative size measures the value of the two firms involved in an M&A
deal, exhibiting a substantial impact on M&A outcomes. Asquith et al. (1983), by examining
the effect of relative size of the target and bidder on bidder gains, noted a very small target firm
relative to the bidder does not increase the value for the bidder. While a target firm whose size
is about half of the bidder generates 1.8% higher abnormal returns for the bidder than the target
whose size is only one-tenth of the bidder. In line with this, Jarrell and Poulsen (1989) argue
that a small target relative to the bidder may fail to raise the market’s attention, leading to
insignificant market reactions to the bid announcement, which is supported by Loderer and
Martin (1990) who include private acquisitions in their M&A sample.
26 These authors adjust for size of the firm in assessing a firm’s long-term stock returns.
Chapter 2. Literature Review: M&A Performance
74
2.3.3.7. Market-to-book ratio
Market-to-book ratio (MTBV) reflects the firm’s prospects, which is also a crucial factor in
assessing M&A performance of the firm. In a core paper, Rau and Vermaelen (1998) find that
the post-merger performance of the glamour bidder (i.e. firms with higher MTBV) is
significantly worse than that of the value bidder (i.e. firms with lower MTBV) regardless of
the choices of mergers and payment methods. The authors attempt to justify this finding with
three hypotheses: the performance extrapolation hypothesis, the method of payment hypothesis
and the Earnings Per Share (EPS) myopia hypothesis.
Analyzing a comprehensive U.S. acquisition comprising 3,169 mergers and 348 tender offers
between 1980 and 1991, Rau and Vermaelen (1998) found that glamour bidders underperform
value bidders in a three-year period following acquisitions. Their study showed that bidders in
mergers underperformed their match firms significantly by 4 percent while bidders in tender
offers enjoyed positive abnormal returns of 9 percent. These results suggest that the market
tends to over-extrapolate the past performance of the glamour firms. The market believes that
firms with high MTBV tend to have high past stock returns and high growth in cash flows and
earnings, leading to an overestimation of the firms’ performance around the announcement
date. It is suggested that the market belief to the firm’s performance enhances managerial
confidence and result in an overpayment, which leads to the same prediction of the hubris
management hypothesis. In addition, the authors revealed that glamour bidders prefer stocks
to cash as a means of payment for acquisitions, as stocks are overvalued by the market in the
presence of the market’s over-extrapolation bias. In the long term, market correction to
overvalued stocks downgrades the price, leading to negative abnormal returns. However, their
results do not provide any additional support for the prediction that the EPS is an important
determinant for the glamour bidders’ underperformance.
Chapter 2. Literature Review: M&A Performance
75
Sudarsanam and Mahate (2003) provide similar findings when examining a sample of
successful U.K. acquisitions between 1983 and 1995. They used two proxies for glamour/value
stocks: price-to-earnings ratio (PER) and MTBV and assessed the firm’s performance with
various benchmark models. Their results showed that the value bidders outperformed the
glamour bidders over three years in the post-merger period, arguing that underperformance in
the long term is due to the fact that the U.K. market forbids the market to over-extrapolate the
firm’s past performance.
Pachare (2010) explains the extrapolation phenomenon with the level of market power that a
firm possesses. He proposes that higher market power possessed by the firm relative to its
industry peers is associated with a stronger positive investors’ belief in the firm’s prospects.
Their study indicates that firms with high market power tends to increase the market’s over-
extrapolation, leading a stronger short-term momentum followed by stronger long-term
reversals.
On the whole, the studies cited above show that MTBV should be included in assessing the
M&A performance of the firm for both the short and long terms. Fama and French (1993),
criticizing the work of Agrawal et al. (1992), who do not adjust for MTBV while assessing the
firm’s M&A performance suggest that MTBV is able to capture a large proportion of the
average stock returns for the firm. Therefore, it is essential to include MTBV in an M&A study.
Chapter 2. Literature Review: Prospect theory
76
2.4. Prospect theory
Kahneman and Tversky (1979) were the first to propose prospect theory, which is a descriptive
theory of decision-making under risks. Such a decision-choice model decipts an S-shaped value
function, showing concave for gains and convex for losses. There is a diminishing sensitivity
in the loss and gain domains, and such a trend in loss domain is much steeper than in gain loss.
This leads to an implication that given an equal change in these two domains people tend to be
more sensitive to losses rather than gains, i.e. loss-aversion tendency. Since people care about
the loss more than the gain, they tend to be risk-seeking when experiencing losses and risk-
averse when enjoying gains. However, gauging gain and loss is subjective. The expected
expected utility27 theory (i.e. EUT) judges people’s feelings about the gain and loss with a zero
value, the positive outcome is described as a gain, whereas the negative outcome is regarded
as a loss. And the theory uses weighted average of outcomes to measure the expected value of
the choice.28
One of the leading features of prospect theory is that the gain and loss is determined by a
specific value point, termed as the reference-dependence bias. Even though the outcome is
postive, it can be regarded as a mental loss if it is below the expectation. In addition, unlike
EUT, which potrays the outcome of a decision, prospect theory puts much more emphasis on
a decision-making process, since it outlines how the change of the value affects the outcome
of a decision. Prospect theory has been widely applied to address various fields of study over
the decades. Barberis (2013) summarised three important implications of prospect theory:
diminishing sensitivity, loss-aversion tendency and reference dependence, and these were
derived from two types of evidence: human-related experiments and empirical evidence.
27 The utility is the value placed on outcome. 28 According to the EUT, The expected value of the choice is sum of the product of the probability and outcome
values of each probability.
Chapter 2. Literature Review: Prospect theory
77
A number of studies document empirical evidence in support of prospect theory. Northcraft
and Neale (1987) conducted an experiment on the estimates of house price. They found that
the experienced house agents estimate house prices in relation to a random house-listing price.
Loughran and Ritter (2002), using the discontinuity analysis, found that current CEOs raise
new equity in accordance with the share price at the time they started to charge the firm. Shefrin
and Statman (1985) observed that investors tend to sell winner stocks too early and keep the
loser stocks too long. In a similar vein, Odean (1998), by studying the detailed trading records
of 10,000 accounts from 1987 to 1993, which were collected from a large discount of brokerage
house, found that individual investors were less likely to sell loser stocks to realise paper loss,
indicating that the loss-aversion is a prevalent feature in the stock market.
2.4.1. Diminishing sensitivity
Kahneman and Tversky (1979) suggest diminishing sensitivities in both the loss and gain
domains. According to their experiment, subjects tend to be more sensitive when the
temperature changes from 3 degrees to 6 degrees than if it changes from 13 degrees to 16
degrees. This assumption also applies to the monetary situation where a loss of $10 has more
impact on a person with $100 than on a person with $1000. It is suggested that the change of
the value affects people’s feelings rather than the final states. Finally, the experiment also
documented that people have a loss-aversion tendency insofar as they are reluctant to realise
losses relative to the reference point. It has been explained that experiencing pain makes a
greater impression than experiencing pleasure.
2.4.2. Reference dependence
A key distinguishing feature between prospect theory and expected utility theory is the
reference-dependence. Prospect theory describes gains and losses on a bias of a given specific
Chapter 2. Literature Review: Prospect theory
78
point, which serves as the benchmark. Kahneman (2003) maintained that prospect theory
captures the decision-making process whilst the expected utility theory simply reflects an
outcome of the decision-making process, failing to address the effect of the change of the value
on people’s feelings. The reference-dependence preference also predicts that decision-makers’
personal judgements are subjective to an established level of reference, and they are satisfied
when returns are higher than the reference point and dissatisfied when the returns are lower.
Tversky and Kahneman (1974) conducted a series of human-subject experiments, affirming
that the estimates of participants are influenced by a random number in case of uncertainty.
Subjects were allocated in different groups and asked the percentage of African countries in
the United Nations. With insufficient information about this unfamiliar area, subjects randomly
estimated a number in relation to the arbitrary number provided by the experiment. The median
estimates for the percentage of African countries in the United Nations were 10 for the group
given the number of 25, and 65 for the group given the number of 45. Therefore, Tversky and
Kahneman identified the reference point effect on the decision-making process.
2.4.3. Loss-aversion tendency
Loss-aversion theory posits that investors generally have a tendency to not be willing to realise
loss (Kahneman and Tversky 1979), where the loss is relative to a reference point. In addition,
Tversky and Kahneman (1992) indicate a diminishing sensitivity suggesting the utility for
returns diminish in both gain and loss domains. The loss-aversion tendency has entered trading
behaviour as a disposition effect. Prior studies have put forward loss aversion theory in
different fields of finance. Using a realised gain and lose model, Barberis and Xiong (2009)
indicate the disposition effect to describe investors’ trading behaviour. Mehra and Prescott
Chapter 2. Literature Review: Prospect theory
79
(1985) and Benartzi and Thaler (1995) suggest that the disposition effect should provide an
explanation for the equity premium puzzle.29
2.4.4. Prospect theory and M&As
Baker et al. (2012) were the first to put forward the idea of applying reference point theory to
M&As. Studying a comprehensive sample of U.S. public acquisitions, the authors found that
the reference point price measured with the target 52-week high, has a substantial impact on
M&A participants’ decisions-making process. Their study indicates that the target 52-week
high serves as bargaining power for the target who negotiates with the bidder. Since the value
of the firm is hard to estimate and lacking time in which to process the information, target
shareholders tend to believe the target 52-week is their firm’s fair value and approve the deals
if offer price is paid according to this reference point price. Target managers who avoid being
accused of failure to meet the commitment to the firm attempt to justify the offer price with
this reference point price. This is similar to bidders who have to justify how much they should
pay for the targets to obtain the deal. According to Baker et al. (2012), the bidder may reason
that ‘if the target was valued at a certain level just a few months ago, shouldn’t we, with our
ability to realise synergies, value it near or above the same level?’, implying that the M&A
pricing decisions made is not only influenced by the reference point but also by the managerial
overconfidence, since bidders do not have any solid evidence regarding the target’s value.
Baker et al’s study (2012) showed that the target 52-week high is significantly positively
related to the offer premium and bidders receive negative market reactions around the takeover
announcement date when paying for acquisitions based on the target 52-week high. The authors
29 Benartzi and Thaler (1995), building on the study of Mehra and Prescott (1985), proposed the myopia loss-
aversion as a solution to the equity premium puzzle.
Chapter 2. Literature Review: Prospect theory
80
reason that targets are less likely to give up firms unless the offer premium can compensate
their mental loss, measured relative to a reference point price. The bidder whose aim is to
successfully obtain the deal is likely to pay a price relative to the target 52-week high to avoid
the loss-aversion of the target. According to the analysis of Baker et al. (2012) regarding the
relationship between deal success and the target 52-week high, it was found that paying
according to the target reference point increases the likelihood of deal succeeding. While, on
the other hand, bidder shareholders find a price relative to the target 52-week high as an
overpayment, since high offer premiums transfer their wealth to that of the target shareholders
and lead the managers hard to recover following the transaction, thus negatively react to the
bid announcement. Their findings are consistent with implications of prospect theory.
Following Baker et al’s idea (2012), Chira and Madura (2015) focused on takeover probability
and the reference point effect. Chira and Madura (2015) suggest that bidders should also assess
their value according to the bidder reference points, finding that optimistic managers who are
more likely to acquire their firms (i.e. a management buyout) than the outsiders when the firm’s
price is low relative to the 52-week high. In addition, their study documented that a firm with
the nearness 52-week high is probably the bidder, whereas a firm with the farness 52-week
high is probably the target in an M&A deal, indicating the role of the firm’s 52-week high
played as the firm’s bargaining power.
81
CHAPTER 3: REFERENCE POINT
THEORY ON CROSS-BORDER AND
DOMESTIC M&A DEALS: U.K.
EVIDENCE
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
82
Abstract
There is no consensus on how much a target is worth except there are perceptions of the manager and
the market to the firm’s valuation. This chapter examines the reference point theory of M&As, a
behavioural finance theory that aims to explain what determines the decision of M&A pricing. The
researcher put forward Baker et al.’s idea (2012) of finding a strong reference point effect on various
key aspects of U.S. M&As in the U.K. market, where the reference point effect is expected to have
stronger explanatory power for offer premiums and the market reactions due to the unique features
of this market. In this chapter, the reference point effect was used to analyse offer premiums and
bidder announcement returns of a sample of 155 cross-border and 451 domestic acquisitions into the
United Kingdom. Using the target firm’s 52-week high as the reference price, an overall significantly
positive relationship between the reference price and offer premiums was established. In addition, a
reference point distinguishes the takeover motives of domestic bidders from cross-border bidders. It
was established that offer premiums driven by the target 52-week high lead to negative market
reactions for domestic bidders, suggesting evidence of overpayment. However, there is little evidence
of overpayment in cross-border bidders who pay in accordance with the target 52-week high. The
findings of this chapter are consistent with prospect theory.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
83
3.1. Introduction
“Necessity never made a good bargain.”
Benjamin Franklin
Valuating a target firm is a difficult task for bidders. A sizable stream of M&A research has explained
the price that a bidder offers for the target from various perspectives of the bidder firm, such as the
synergy hypothesis and the overpayment hypothesis (Moeller et al., 2004; Alexandridis et al., 2012).
Under the synergy hypothesis, offer premiums can be viewed as an exchange of the expected
synergies following acquisitions (Antoniou, et al., 2008), while under the overpayment hypothesis,
offer premiums are a result of overpayments, which are largely because of managerial overconfidence
(Hayward and Hambrick, 1997; Malmendier and Tate, 2005).
The price that the bidder offers to the target is an outcome of the negotiation between the boards of
the target and the bidder, and such a price involves the perception of the managerial view of the firm’s
valuation. Since how large the M&A offer premiums are paid for the target determines the wealth
distribution between the target and the bidder shareholders, an offer cannot be made without an
agreement between the two firms involved. With this in mind, analysing takeover motives from either
the bidder or the target side alone does not fully account for the rationale behind the offer price.
In this chapter, the reference point theory of M&As was established to explain the offer price. The
theory was first developed by Kahneman and Tversky (1979), depicting how people make decisions
under risks. Putting this in the context of M&As, the theory sheds light on how managers make major
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
84
investment decisions facing complex situations. Apart from those behaviour finance theories that
explain M&A motives from the outcome of the deal, prospect theory allows use to directly observe
the managerial decision-making process of the offer premium.
Edmister and Walkling (1985) suggest that the offer premium reflects the bargaining power of the
management teams of the target and the bidder, implying that valuing a firm requires a series of
cognitive biases of the management team of the bidder and the target. Due to the complexity of the
M&A activity, the offer premium cannot be set precisely. Therefore, addressing the M&A valuation
effect requires further investigation from the perspective of managerial cognitive biases.
The reference point effect is an important implication of prospect theory developed by Kahneman
and Tversky (1979), affirming that people have a pre-determined piece of information in mind and
gauge losses and gains according to it, and people’s feelings about the loss and gain tend to diminish
according to the reference point. They are more sensitive to the value that is nearer the reference point
than far from it (i.e. diminishing sensitivities). In addition, prospect theory posits that people are
generally loss-aversion, in that they are not willing to realise loss relative to the reference point they
have settled upon. The implications of the reference point theory have been highlighted in
psychological research (Tversky and Kahneman, 1974) and also applied in financial and economic
research (Shefrin and Statman, 1985; Odean, 1998; Barberis and Xiong, 2009).
A firm’s 52-week high is publicly available information and serves as a salient reference price in
previous literature (Heath et al., 1999; Huddart, et al., 2009; Burghof and Prothmann, 2011; Baker et
al., 2012). In Tversky and Kahneman (1974), people tend to rely heavily on a single piece of
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
85
information while making decisions. The 52-week high has been widely reported by some financial
media, which shapes investors’ impression. In addition, investors are likely to be influenced while
lacking information. Burghof and Prothmann (2011) suggest that the 52-week high effect is more
pronounced when there is high level of information uncertainty.
Baker et al. (2012) proposed the reference point theory for M&As, examining how target valuation
could alter the offer premium. These authors were the first to put forward the idea of applying the
reference point effect on an M&A testing ground. The theory explains the offer premium from a
psychological perspective from both the bidder and the target sides. The authors indicate that the
reference point measured by the target 52-week high is positively related to the offer premium. They
find that bidders tend to offer a price for their targets relative to the target 52-week high, which
suggests that both targets and bidders are influenced by the reference dependence bias (i.e. the
reference point effect). In addition, it was found that the market tends to react negatively to the bidder
announcement when the offer price is paid according to the target 52-week high, which implies that
the target 52-week high is employed to reinforce the bargaining position of the target during the
course of the deal negotiation. If a high offer price destroys the wealth of bidder shareholders, why
bidders pay in accordance with the target 52-week high. On the other hand, what facilities the target
in using the reference point requires further examination.
This chapter provides further evidence in relation to the reference point effect on the offer premium
the U.K. market. The reference point effect in a comprehensive public acquisition sample of both
U.K. domestic bidders and cross-border bidders into U.K. public targets will be investigated. A
comprehensive public acquisition sample of both domestic and cross-border will enable assessment
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
86
of the role of the reference point effect plays in a different setting outside U.S. market. Previous
studies show the level of bidding competition is more intensive in the U.K. takeover market than any
other takeover market in the world (Moeller and Schlingemann, 2005). Alexandridis et al. (2010)
found that higher level of competition is accompanied by positive target announcement returns,
higher offer premiums and negative bidder announcement returns. Their findings suggest that public
targets in a more competitive market have relatively greater bargaining power in M&A activity, and
are likely to translate it into higher offer premiums based on their 52-week high. Targets in such an
environment are more likely to use a reference point price to negotiate for a possible high price from
the bidder. Therefore, the researcher hypothesises that the reference point theory for M&As is
pronounced in the U.K. market, which is also competitive.
The researcher contributes to this thread of literature by directly examining the role of reference point
theory in the setting of cross-border M&As of U.K. targets. Erel et al. (2012) and Uysal et al. (2008)
affirm that the likelihood of acquisitions increases when the target nation is closer to the bidder nation,
which suggests geographical distance impedes information transmission in matters of investments.
Domestic bidders are supposed to have informational advantages over cross-border bidders due to
the sophisticated local networks. Therefore, it is expected that cross-border bidders lacking
information about their targets are more likely to put more weight behind publicly accessible
information such as the target 52-week high. Therefore, bidders might have to offer a premium based
on the target 52-week high.30
30 Croci et al. (2012) provided an alternative explanation regarding information asymmetry and offer premiums, arguing
that bidders with favourable asymmetry information tend to offer higher premiums. Bidders can therefore be expected to
regard the target 52-week high as its potential profit-generating ability, and offer high premiums when the target 52-week
high supports the favourable asymmetry information they might have.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
87
Furthermore, strong investor protection in the U.K. market could potentially reinforce the reference
point effect. Under the shareholder-oriented legislation environment, U.K. target managers are
unlikely to pursue private benefits at shareholders’ expense (Armour and Skeel, 2007). Wulf (2004),
however, found U.S. evidence that agency conflicts between the target manager and the shareholder
result in low offer premiums as target managers accept cheap offers associated with larger personal
benefits. Therefore, the reference point effect should be more pronounced in markets with stronger
investor protection than those without. In this pursuit, the U.K. market also provides an ideal testing
ground for the reference point theory.
Accounting for the majority of the deals in the European countries, the U.K. takeover market is the
second largest market in the world (Faccio and Masulis, 2005; Croci et al., 2010; Gugler et al.,
2012).31 Recent decades have seen a rapid growth of cross-border bidders’ acquisitions of U.K. public
targets. Many additional studies focus on the cross-border M&As into the U.K. market. Danbolt (1995)
attributes the U.K. takeover boom in the late 1980s to the growth of cross-border acquisitions.
Following his study, Gregory and O'Donohoe (2014) analysed 119 cross-border bidders into the
United Kingdom over the period 1990-2005, where the cross-border acquisitions took up to 41% of
their U.K. public acquisition sample.
31 By studying comprehensive European acquisitions announced between 1997 and 2000, Faccio and Masulis (2005)
reported a large proportion of deals involving U.K. bidders (65.3%) and U.K. targets (47.0%). Croci et al. (2010) indicate
that U.K. takeover market is an active takeover market which accounts for 65% of mergers in Europe. By studying the
U.K. completed mergers during the fifth merger wave, Gugler et al. (2012) found that the U.K. experiences a similar
takeover trend compared with the United States. In the U.K. M&A market, the overall transaction value reached $360
million in 2000. At the same time, the total transaction value for the M&A market in other major European countries was
$260 million.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
88
Given the strong investor protection in the U.K. market, cross-border M&As directed at U.K. targets
provide a natural hedge against any economic and political instability for the bidder. To the best
knowledge of this researcher, this is the first study using a cross-border M&A sample to test the
reference point theory of M&As in the U.K. market. The findings of this chapter imply that stronger
investor protection enhances the target bargaining power, increasing the probability of the target
requires higher offer premiums from the bidder based on the target 52-week high.
The results of this chapter are consistent with the implications of prospect theory. This chapter yields
two main findings. First, it became apparent that the target 52-week high is significantly positive
relative to the offer premiums, suggesting that an offer price linked to the target 52-week high is
perceived as a reference point for the target shareholders in both domestic and cross-border M&As.
With the prospect theory framework, targets are generally loss-aversion, in that they are reluctant to
accept an offer that is significantly below the reference point, hence pushing their managers to
negotiate for higher offer premiums. Targets find it easier to require high offer premiums from the
cross-border bidders than from the domestic bidders. However, it was found that bidders no longer
pay based on the target 52-week high when the current price deviates greatly from it. It is possible
that the target’s current price is less likely to reflect its best performance in the recent past, leading
bidders to walk away. Our results therefore imply that bidders are bounded by rationality.32
Secondly, the market reacts negatively to the bid announcement in the sample consisting of both
domestic and cross-border M&As, which suggests that the market views an offer price based on the
32 Despite the fact that bidders make decisions based on the superficial information they possess they have rationality of
preventing things from going wrong.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
89
target 52-week high as an overpayment. While investigating the role of the reference point effect in
these two subsamples and using the target 52-week high as an instrumental variable of the offer
premium, it was established that the offer premium driven by the reference point effect does not make
any greater difference to the bidder announcement returns under the 2SLS estimator than by OLS in
the cross-border M&A sample. It does, however, lead to more negative bidder announcement returns
under 2SLS in the domestic M&A sample. These results show in the case of domestic bidders who
offer a price according to the target 52-week high tends to be perceived as overpayment, whereas the
market has already taken the reference point effect into consideration for bidders making cross-border
M&As.
Two distinct contributions have been made to M&A literature. First, to the best of the researcher’s
knowledge, this is the first study to test the reference point effect using cross-border M&A sample.
The cross-border M&A sample was used as a control sample for the domestic M&A sample to
investigate the reference point effect in a setting where there exists larger information disadvantages.
It is expected that the reference point effect more pronounced among cross-border bidders given they
have a less sophisticated information networking than their domestic counterparts. Therefore, the
reference point theory helps explain how bidders value targets in an environment of information
uncertainty.
Second, the reference point theory distinguishes the motive of domestic bidders from cross-border
bidders acquiring U.K. public targets. The offer premium based on the reference point has a
significantly negative impact on domestic bidders’ announcement returns yet an insignificant impact
on cross-border bidders’ announcement returns. It can be argued that shareholders of domestic
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
90
bidders may believe that managers pay too much for the target, based on the target 52-week high, and
thus react negatively to the bidder announcement, whilst those of cross-border bidders may take the
target 52-week high for granted.
The remainder of this chapter is organised as follows: Section 2 provides a literature review related
to the offer premium, prospect theory, and cross-border acquisitions. Section 3 develops hypotheses
with regard to the reference point effect. Section 4 summarises our data and presents the methodology
used in this chapter. Section 5 analyses our results. Section 6 concludes this chapter.
3.2. Literature review
3.2.1. Cross-border M&As in the U.K.
The level of cross-border acquisitions in the U.K. has grown rapidly over the past three decades.
Danbolt (1995) held that the U.K. takeover boom during the 1990s was largely because of the increase
of cross-border acquisitions. His study documents that cross-border acquisitions into the U.K.
accounted for 58% of the total value of this market in 1990. Erel et al. (2012) conducted a study into
a sample of 56,978 cross-border acquisitions in 48 countries between 1990 and 2007, taking up one
third of worldwide acquisitions. It was found that the volume of cross-border acquisitions increased
from 2002 to 2007, reaching to 45% of all M&A activities up to 2007. Of particular note, U.K. targets
were engaged in 6,753 cross-border acquisitions.
The level of U.K. M&A activity has turned the market into an active takeover market. Faccio and
Masulis (2005) in a sample of European M&As over the period of 1997 to 2000 reported that a 47%
of acquisitions involved U.K. targets. Consistent with their study, Croci et al. (2010) found that the
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
91
level of U.K. M&As took up 65% of takeover activities in Europe between 1990 and 2005, suggesting
that U.K. is the most active takeover market in Europe. Moeller and Schlingemann (2005) argue that
the U.K. takeover market was even more active than the U.S. for its strong shareholder right
protection and strict legislative environment. Alexandridis et al. (2010) ascribe the activity of the U.K.
takeover market to intense competition. Their study shows that the number of competing bids in the
U.K. was more than that in the United States.
Research has identified the motives of cross-border acquisitions. For example, Danbolt (2004) noted
that cross-border acquisitions may not be simply driven by value maximisation as to the same degree
as domestic acquisitions. Instead, bidder managers are prone to increase their power, status and salary
through cross-border acquisitions. Doukas (1995) and Doukas and Travlos (1988) argue that cross-
border acquisitions play a value-enhancing role as they integrate the imperfect capital markets across
countries, while Erel et al. (2012) noted that cross-border acquisitions were driven by the high quality
of governance standards. Their worldwide evidence shows that in countries with a higher quality of
corporate governance, accounting standards have stronger shareholder protection for the combined
entities. Rossi and Volpin (2004) provided some similar findings, showing that the volume of M&A
activity is significantly larger in countries with better accounting standards and stronger shareholder
protection. Their findings imply that cross-border acquisitions act as a tool for a country with a
substandard legislative environment to access a country with a better legal environment.
One particular U.K. study by Danbolt (2004) identified the international risk diversification
hypothesis as the most plausible explanation regarding the motive of cross-border acquisitions in the
the U.K. compared with the market access hypothesis and the exchange rate hypothesis. Based on a
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
92
sample of 514 domestic and 116 cross-border acquisitions in the U.K. setting over the period between
1986 and 1991, Danbolt found that targets gained significantly higher abnormal returns in the cross-
border acquisitions than in the domestic acquisitions. In addition, his study shows that geographical
distance between the bidder country and the target country affects the wealth effect of the target
shareholder. The target abnormal returns were found to be significantly higher when bidders are based
outside the U.K. market. However, his results were not robust when different event windows were
employed in his study.
Despite the fact that substantial research has identified the benefits brought by cross-border
acquisitions, valuing a target is a more difficult task for cross-border bidder managers than for
domestic bidder managers given different accounting standards, cultures and a large amount of
asymmetric information country by country. Danbolt (2004) noted that cross-border bidders were
prone to pay higher offer premiums for their targets due to the size of valuation error caused by these
differences between countries.
Uysal et al. (2008) in their study of the relationship between geography and bidder returns in the U.S.
context found that bidders tended to gain higher returns when acquiring a closer rather than a farther
target, which is because of the information advantage obtained from geographical proximity. The
authors argue that information is likely to be coded, transmitted, and interpreted in a nearer country
rather than a more distant one. In addition, local managers are likely to obtain information through
their social networks.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
93
By studying an international cross-border acquisition sample, Erel et al. (2012) found the likelihood
of acquisitions increases when a bidder country is closer to its target country. Their results show
significantly positive coefficients of geographical proximity at a 1% level to explain the bidder returns.
Together, bidders engaging in cross-border acquisitions may encounter information barriers that
potentially cost the bidder and benefit the target.
3.2.2. The reference point effect on the U.K.
While most work has focused on the U.S. market, it is interesting to explore the reference point effect
on cross-border acquisitions in the U.K. public targets. First, the high level of competitiveness in the
U.K. market entitles its targets to greater bargaining power, which makes the reference point effect
pronounced. Danbolt (2004) affirm that the cross-border target effect means that U.K. targets are
outperformed in cross-border acquisitions rather than in domestic counterparts due to risk
diversification. In line with his work, Alexandridis et al. (2010), by investigating a sample of global
M&A activity over the period of 1990 and 2007, found that the level of competition increases with
target abnormal returns and the offer premium, and decreases with bidder abnormal returns. Their
results show that when the market is more competitive, targets are more likely to translate their
bargaining power into benefits. It is generally believed that cross-border bidders have less information
advantage than domestic bidders, which reduces their bargaining power over their targets. With this
in mind, U.K. targets are more likely to ask for an offer price based on a reference price favouring
their shareholders from cross-border bidders than domestic bidders.
Secondly, less severe of agency conflicts between the manager and the shareholder in the U.K. could
potentially enhance the reference point effect. Although it is generally believed that the corporate
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
94
governance system in the U.K. is consistent with the U.S. (La Porta et al. 2000), yet the U.K. has a
better corporate governance system in relation to M&As compared with the US. Armour and Skeel
(2007) in their review of U.S. and U.K. takeover regulations conclude that the U.K. has self-regulation
takeover regulations which are strongly orientated towards the shareholder’ interests, whereas the
U.S. law system entitles managers to absolute discretionary power over the firms. In addition, in the
context of takeover defences the authors suggest that the U.K. takeover code implicitly bans any
“frustrating actions” against their shareholders, which is different from the United States. Thus,
takeover regulations in the U.K. can potentially reduce the conflicts of interests between the manager
and the shareholder. The bargaining power of target managers is enhanced under more stringent
regulations and target managers can require higher offer premiums from outward bidder managers in
M&As.
The corporate governance system is firmly rooted in the legal environment of a country. Research
has established some contradictory findings in relation to the divergence of the legal environment
effect on shareholder valuation in cross-border acquisitions. Moeller and Schlingemann (2005)
analysed bidders’ acquisitions of the worldwide targets over the period of 1985 and 1995 and found
that bidders tend to pay higher offer premiums to targets with weaker legal environment compared
with the United States. Thus, targets with more information asymmetry and poorer corporate
governance system tend to accept lower offer premiums, which leaves the bidders opportunities to
explore the benefits.33
33 Addressing the U.K. market, Antoniou et al. (2008) noted that the lower offer premium does not benefit bidders.
However, their study does not include cross-border acquisitions.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
95
Hagendorff et al. (2008), analysing the bidder returns of financial firms between 1996 and 2004 in
the context of the U.S. and Europe, found that target countries with a weaker legal environment are
more likely to accept fewer offer premiums. Unlike Moeller and Schlingemann (2005), Hagendorff
et al. (2008) argued that information asymmetry and severe agency problems lead to a decrease in
the offer premium.
Martynova and Renneboog (2008) studied a sample of listed bidders’ acquisitions of both listed and
unlisted targets in Europe during the fifth M&A wave. They revealed bidders obtain positive returns
when acquiring targets regardless of their legal environment. They ascribed targets in stronger legal
environment compared with bidders to the bootstrapping effect that bidders have incentives to get
access to better corporate governance system. Based on their study, targets in a strict legal
environment are expected to have greater bargaining power than those in a less strict legal
environment.
On the whole, studies have highlighted how the legal environment impacts the bargaining power of
both the bidder and the target in the context of M&As. However, results are mostly driven in the U.S.
or the European sample in different sample periods (Moeller and Schlingemann, 2005; Hagendorff et
al., 2008, Martynova and Renneboog, 2008). Little has been done relating to the issue of the U.K.
market. Given that the U.K. is a more active and competitive market than that of any other country in
the world due to the stringent shareholder protection and legal environment (Moeller and
Schlingemann, 2005; Alexandridis et al., 2010), target managers are entitled to greater bargaining
power and tend to demand higher offer premiums accordingly. Bidders from countries with relatively
a weak legal environment compared with the U.K. are more likely to pay higher offer premiums in
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
96
exchange for stronger shareholder protection. With a higher level of information asymmetry, bidders
are more likely to be misled by the value of the target, thus valuing targets based on readily available
information. Results regarding the reference point effect being more pronounced on cross-border
M&As.
3.2.3. The target 52-week high as a reference point price
Over the years, attention has focused on a firm’s 52-week high as a reference price. The 52-week
high is widely reported by a range of media on a daily basis, such as the Wall Street Journal, Yahoo
Finance, Financial Times, or the South China Morning Post, leading investors to gain full access to
information and may use it to assess the firm’s performance.
Prior research has identified the 52-week high’s salient role as the reference point in the decision-
making processes. Baker et al. (2012) applied the reference point theory in a comprehensive sample
of U.S. public acquisitions, documenting that the reference price measured with the target 52-week
high had a substantial impact on M&A decisions-making process. The target 52-week high is found
to be significantly positively related to the offer premium and bidders receive a negative market
reaction around the takeover announcement date when paying for acquisitions based on the target 52-
week high. The authors explained it with the reference point theory, reasoning that targets are less
likely to give up firms unless the offer premium can compensate target shareholders’ mental loss
relative to the reference price. Their statement implies the reason that bidders pay in accordance with
the target 52-week high is to avoid target shareholders’ loss-aversion, and thus increase the likelihood
of deal completion. It was found that shareholders negatively react to the bid announcement since
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
97
they believe paying based on the target 52-week high is too high. Baker et al’s findings (2012) are
consistent with the implications of prospect theory.
The 52-week high is significant in explaining investors’ behaviour in a variety of studies. Heath et al.
(1999), for example, found a relationship between individual stock option exercise decisions and the
reference point. Their study includes stock option exercise decisions made by over 50,000 employees
at seven publicly traded corporations and finds that the number of employee stock option exercise
activities nearly doubles when the stock price exceeds the maximum price of the previous year. The
authors reason that individual investors have a reference price relative to the maximum stock price
when exercising stock options.
Huddart et al. (2009) provided evidence in a large sample that there is high trading volume around
the past price extremes. Using the weekly observations of 2,000 firms spanning 24 years from 1982
to 2006, the authors found that the past price extremes are significantly related to the trading volume
at the 1% level. The results continue to hold after controlling for a series of variables that have
significant impact on the trading volume addressed in earlier studies. The authors attribute the
salience of the past price extremes to bounded rationality and attention hypotheses that individual
investors tend to rely heavily on past price extremes, since they are limited in obtaining and analysing
information. Their study complements evidence that the 52-week high has a strong effect on the
trading decisions of the investor.
In a similar vein, Burghof and Prothmann (2011) suggest that the 52-week high effect can be
explained by the anchoring bias hypothesis, affirming that the 52-week high has strong predictive
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
98
power when there is a high level of information uncertainty. Investors can obtain profits under a
greater level of information uncertainty by focusing on the 52-week high strategy. Their results
indicate significant reference point effect on the trading behaviour of the investor.
The use of a certain piece of information as a reference point has been well documented in Tversky
and Kahneman's work (1974). It has been well known that human minds are likely to focus on a single
event and overlook other events that lead to different consequences when making decisions. Their
concept has been put forwarded into various studies and the firm’s 52 week high has been widely
adopted as a reference point candidate for its simplicity. Following these studies, the investor’s
decisions in conjunction with the firm’s 52-week high were measured.
Following Tversky and Kahneman’s work (1974), Rau and Vermaelen (1998) suggest the over-
extrapolation hypothesis indicating that investors put considerable weight behind the past
performance of the firm and attribute a higher expectation to the firm’s M&A performance when it
achieves a decent past performance, whilst they may underreact to a firm with a worse past
performance. Their study suggests that investors have a tendency to make predictions by exploring
the firm’s past performance. Based on the assumption that the market is not efficient, investors’ biases
on the past price information are likely to come into play in estimating the firm’s future performance.
3.2.4. M&A offer premiums
M&A offer premiums are an important factor that have received great deal of attention from
shareholders, as it directly determines how the value of the deal is distributed between the bidder and
the target shareholders. Bidders who pay high offer premiums for the target firms tend to signal to
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
99
their short-horizon shareholders that the firm is hard to generate wealth in the short run even though
it is possible for the firm to create value in the long term. On the other hand, target shareholders
benefit from high offer premiums. With this in mind, a higher offer premium is generally associated
with negative market reactions to the bidder firms but with positive market reactions to the target
firms. Prior literature on the offer premium has largely focused on the two completing hypotheses:
the synergy and the overpayment hypotheses. The synergy hypothesis tends to rationalise the motive
of bidders paying for the target firm as being to create value for the firm, suggesting that managers
serve the best interests of their shareholders should finding positive NPV projects, and premiums paid
for a target are in exchange for potential earnings to be created by a combined entity. The
overpayment hypothesis suggests that bidders are overconfident or the agency problems of the firm
are severe, leading managers to offer the target higher premiums in exchange for the control of the
target firm.
To justify the motive of offer premiums, Berkovitch and Narayanan (1993) tested three hypotheses
of M&A motives in a U.S. study. Their work provided a way of distinguishing synergy from
overpayment by looking at the correlation between target and total gains. It is suggested that
correlation should be positive if the bid is solely driven by synergy but negative if the bid is solely
driven by agency problem, and zero if the bid is solely driven by hubris. Their results indicate that
synergy and agency problem primarily explain the motives of U.S. M&As, with which Hodgkinson
and Partington (2008), agreed in their study of a sample of U.K. M&As over both the short- and long-
term windows, finding that bids are largely motivated by synergy and hubris, and that there is weak
evidence of agency-motivated bids.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
100
Re-addressing the offer premium in the U.K. market, Antoniou et al. (2008) found that the synergy
hypothesis has more explanatory power than the overpayment hypothesis. Their study is based on a
sample of 396 successful U.K. public acquisitions between 1985 and 2004. The authors divide the
bidder sample into the higher, medium, and lower offer premium sub-portfolios, and conduct both
the short- and long-term analysis for each sub-portfolio. It was found that the short-term combined
abnormal returns (i.e. synergy) were positively correlated with the offer premium, and the long-term
abnormal returns to bidders did not show any significant differences between the higher and the lower
offer premium sub-portfolios. It is therefore suggested that synergy is a more plausible motive than
overpayment in explaining the offer premium in the United Kingdom.
Another area of literature addresses the offer premium with the overpayment hypothesis. In earlier
research, Roll (1986) found that hubris-infected managers are engaged in value-destroying bids.
Reviewing a series of U.S. and U.K. studies, the author concludes that managers are accused of
overpaying targets. They tend to believe that they have better-than-average skills which lead them to
dominate the market. As a matter of fact, neither the bidder nor the target managers can make correct
predictions as to a firm’s performance around the takeover announcement date. Their results show
that the combined entity generates significantly negative abnormal returns following acquisitions,
reasoning that managers are irrational and the market is efficient, thus managers are unable to explore
mispricing of the market and thus overpaying for their targets due to an overestimation of the expected
synergies.
In the light of this, Hayward and Hambrick (1997) investigated a sample of 106 large acquisitions
and found evidence of hubris-infected CEOs. These authors focused on large acquisitions
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
101
representing important decisions for the firms, tending to expose their CEOs to extensive risks. Their
study shows that bidders who achieve a decent performance in the recent past receive praise from the
media, together with higher salaries from their firms, which become the most important factors
leading them to believe that they are important in the firm. Thus, they are affected by hubris while
making major investment decisions. Their study suggests that the bidder’s recent performance, the
recent media praise for the CEOs, the CEO’s self-importance, and the composite index of these three
factors are positively correlated with the size of the offer premium, incurring negative market reaction.
Hayward and Hambrick's findings (1997) suggest that higher offer premiums are paid by hubris-
infected CEOs.
Using a comprehensive sample of U.S. M&As over the period of 1980-2001, Moeller et al. (2004)
established a positive relationship between the offer premium and the value of the bidder firm. The
authors provided various plausible reasons to explain why larger bidders tend to overpay for their
targets than their smaller bidders. First, larger firms generally have less concerned ownership than
smaller firms, leading to a weaker link in the interests between the manager and the shareholder.
Secondly, managers in a large firm are likely to be more overconfident as they are able to allocate
resources more easily, without any obstacles. Thirdly, larger firms are more constrained in finding
growth opportunities than smaller firms, thus managers are more likely to dissipate resources in
unnecessary projects, which destroy the wealth of their shareholders. Their findings indicate that
bidder managers of a larger firm, due to greater autonomy, are likely to be more overconfident than
those of a smaller firm.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
102
However, Alexandridis et al. (2013) found little evidence that value-destroying acquisitions are
caused by large M&A offer premiums to the target shareholders and reason that the negative source
of the firm’s value is the neglect of the M&A integration process. They find an overall negative
relationship between the target size and the offer premium. Their study reported that the mean
(median) premium for the larger target is 38% (32%), whereas the mean (median) premium for the
smaller target is 54% (45%). Though a lower offer premium is paid to the larger target, it does not
create value for the firm. The authors assert that larger targets are value destroying for bidders, due
to bidders’ neglect of the target post-merger integration.
Incorporating both the synergy and overpayment hypotheses into the M&A study and using the
quadric model for a sample of 49 banking industry M&As between 1995 and 2004, Díaz et al. (2009)
found both the synergy hypothesis and the overpayment hypothesis come into play in their M&A
sample. More specifically, the authors suggest a non-linear relationship between the offer premium
and bidder abnormal returns. The lower level of the offer premium paid to the target shareholders
tends to create synergies for the firm, whilst the value of the combined entity is destroyed when the
offer premium further increases. Based on these results, the authors suggest the synergy hypothesis
is able to explain the lower level of the offer premium and the overpayment hypothesis is able to
explain the higher level.
Many additional studies have largely explained the offer premium in relation to these two competing
hypotheses. Bowman and Richards (2013) suggest that a higher offer premium is used in exchange
for the market power of the bidder, which generates synergy following acquisitions. Soegiharto (2009)
in his study of 3184 mergers between 1990 and 2010 finds no evidence that CEO hubris affects the
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
103
offer premium. By studying 81 European banking M&A over the period of 1994 to 2000, Diaz and
Azofra (2009) found that bidders tend to pay more for targets that have attractive characteristics, such
as it is larger size and has a decent recent performance. These authors argue that managers can have
substantial private benefits when acquiring another bank successfully, indicating that agency
problems are a source of overpayment. By contrast, Kim et al. (2011) argued that the offer premium
is to reflect upon the bank managers’ willingness for future growth. In that case, there are no conflicts
of interests between the manager and the shareholder and the primary purpose of paying M&A offer
premiums is to exchange for the control of the target firm.
3.2.5. Reference point price and the offer premium
Prospect theory offers new insights into the M&A offer premiums. It addresses the fact that prices
offered to the target shareholders not only originate from market reaction but also provides many
additional considerations to explain the behaviour of the main M&A participant. It differs from those
behavioural finance explanations relating to deal synergies and provides access to the market looking
at the negotiation process of the two firms. Since the offer premium is the outcome of negotiation
between the boards of the two firms involved, it reflects how the two firms think about the deal during
the course of the negotiation. Given a lack of information to observe the negotiation process of the
two firms, a high offer premium may signal to the market that the takeover target tends to dominate
the negotiation table and it is able to demand a price favouring their shareholders. Bidders, on the
other hand, will defend their position by providing market evidence that the target has the potential
to create synergies following an M&A deal, thus rationalising the M&A offer premiums. How much
a bidder firm should offer for the target firm is far too complex, whilst a single salient piece of public
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
104
information is more likely to impress the investors’ minds. By borrowing this, the bidder tends to
rationalise his/her motive for paying for the target and avoids being blamed on the market.
Baker et al. (2012) found that bidders tend to offer the target a price that is based on the target 52-
week high, indicating that bidders believe that they can beat the target 52-week high and thus create
synergies. However, their results show that the offer premium driven by the target 52-week high is
negatively related to the bidder announcement returns. The authors studied the role of the target 52-
week high plays in M&As. By doing so, they examined the OLS regression results for the bidder
announcement returns on the offer premium and compare these with the 2SLS regression results for
the bidder announcement on the offer premium, where the target 52-week high is used as an
instrumental variable of the offer premium. Baker et al. expect less negative or even positive market
reactions to bidders under the 2SLS compared with those by the OLS if the target 52-week high
reflects higher synergies and a more negative or a less positive bidder announcement returns if the
target 52-week high indicates overpayment.
Baker et al. (2012) suggest the role of the target 52-week high is an overpayment. Specifically, the
coefficient of the offer premium predicted by the OLS estimator is -0.040, suggesting that an increase
of 10% for the offer premium leads to a 0.4% decrease in the bidder announcement returns, while
using the target 52-week high as an instrumental variable of the offer premium. So, when using the
2SLS estimator, the coefficient for the offer premium becomes -0.245, suggesting that the target 52-
week high leads to more negative bidder announcement returns. Their results indicate that the market
believes the bidders’ offer to the target, whose current price is far below its 52-week high, is an
overpayment whereas the bidder may believe such a target leaves the firm more room for synergy
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
105
creation. Therefore, the manager and the shareholder tend to interpret this certain piece of price
information differently.
3.3. Hypothesis development
According to prospect theory, people tend to gauge gains and losses relative to a reference point, and
are unwilling to realise mental loss relative to this particular point. Since valuing a target is a difficult
task in M&As, the 52-week high is likely to be used for its simplicity. Burghof and Prothmann (2011)
suggest the use of a reference price increases with the level of information uncertainty. Moreover, the
52-week high reflects a firm’s recent best performance which provides predictions for the future
performance (George and Hwang, 2004). Finally, based on some experimental evidence from
Tversky and Kahneman (1974) and Kahneman and Tversky's works (1979), investors tend to rely on
a piece of information while deciding an atmosphere of risk and uncertainty. In the meantime, the 52-
week high as a reference price has been drawing the attention of the investor through a massive report
by the financial media.
The use of the reference point can be explained by the psychological influence of both the bidder and
the target. Target shareholders may find it hard to gauge managerial performance in M&A activity,
since they do not know for how much exactly their firm should be sold. Lacking information and time
in which to process the information related to the deal, target shareholders are likely to estimate their
firm based on a straightforward relevant price measure, such as the 52-week high. With a strong loss-
aversion tendency, they are not willing to see their firm, sold at a price significantly below their 52-
week high.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
106
The psychological influence among the target shareholders has a substantial impact on the decision-
making processes of their manager. If the target manager sells a firm at a price far below the 52-week
high without a rational motive, the shareholders must blame their managers by arguing that they
would have been better off if they had not sold the firm. Managers thus may incur legal risks. The
magnitude of the premium received by the target manager can be seen as a judgement of whether or
not the manager aims to protect the wealth of their shareholders. It has been shown that U.K. has a
high quality of corporate governance system, which leads to less severe conflicts of interests between
the managers and the shareholders. In this stringent legal environment and shareholder protection,
the wealth of the target shareholders is the first priority. Target managers would bargain for the
premium based on the 52-week high, which is a visible price for the market. Baker et al. (2012)
indicate that bidders tend to pay a price based on the target 52-week high if it is closer to the target
current price, arguing that it might be a true reflection of the real performance of the target. Together,
a positive correlation between offer premiums and the reference price measured by the target 52-
week high in the U.K. market is expected.
H1a: There is a positive correlation between offer premiums and the reference price in the U.K.
market.
According to Erel et al. (2012) and Uysal et al. (2008), cross-border bidders have greater
disadvantages in gathering information than domestic bidders due to geographical distance. As the
level of information asymmetry increases, cross-border bidders are less likely to accurately estimate
the true value of the target, whereas domestic bidders tend to collect exclusive information from their
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
107
local social networks. In such a case, cross-border bidders are more likely to measure firm valuation
based on the target 52-week high.
In addition, a more competitive market environment tends to increase target bargaining power, thus
enhancing the reference point effect. Targets may find it easier to negotiate the offer premium based
on their 52-week high with the cross-border bidders than with the domestic bidders. In Edmister and
Walkling (1985), the offer premium is the outcome of negotiation. In the light of this, targets could
require higher premiums when they have greater bargaining power. Therefore, a stronger reference
point effect on cross-border acquisitions toward U.K. targets is expected.
H1b: There is a positive correlation between offer premiums and the reference price when cross-
border bidders acquire U.K. targets.
Moeller et al. (2004) suggest that overpayment increases the wealth of the target shareholder and
decreases the wealth of the bidder shareholder. The target 52-week high is perceived as the recent
best performance of the firm. Bidder managers are expected to possibly a lower price so as to increase
the wealth of their shareholders. Bidder shareholders may simply believe their managers pay too
much if their payment is based on the target recent best performance. It is generally believed that an
overpayment leads to a negative market reaction since bidder shareholders may believe such
premiums paid to the target are hard to recover following acquisitions. There, bidders’ performance
around the announcement date are expected to be negatively correlated to the offer premium when
deals are made with regard to the reference price.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
108
H2: Bidder short-run performance is negatively correlated to offer premiums when deals are made
with regard to the reference price.
3.4. Data and methodology
3.4.1. Data
The initial sample contained 4,324 acquisitions of U.K. public firms announced between January 1,
1985, and December 31, 2014, extracted from Thomson One Mergers and Acquisitions Database.
The Database contains deal-related information, including the deal number, the DataStream Code of
the firm, which is used to match a firm’s accounting data from DataStream, the transaction value, the
shares percentage acquired by the bidder during the course of the transaction and share percentage
the bidder owned after the transaction, which will aid in determining whether a firm’s control has
been transferred in a transaction, the payment method, deal choice, deal type, deal attitude, SIC code
of the firm from which it can be judged whether or not the deal is diversified.
The sample involved 3,078 U.K. domestic acquisitions and 1,246 cross-border acquisitions into the
United Kingdom. Deals with a missing offer premium has been cleaned, yielding 1,826 acquisitions.
Observations with missing value of bidder 5-day CARs around the announcement date were excluded,
which left a sample of 1,435 acquisitions.34 Acquisitions with the information of payment method
were required, which left a sample of 1,212 acquisitions. Variables of both bidder and target firm
characteristics used in the regressions were not a missing value, which resulted in a final sample of
606 acquisitions, with 451 domestic acquisitions and 155 cross-border acquisitions.
34 The sample size reduces mostly because of bidders from countries without stock market returns, which is in line with
Erel et al. (2012).
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
109
Table 3.1 depicts the summary statistics for a sample of 606 public acquisitions studied in this chapter.
The mean value for the deal was $606.4 million. Of these 606 acquisitions, 380 were all-cash
acquisitions, 108 were all-stock acquisitions. There were 292 diversified acquisitions, 25 hostile
acquisitions, and 408 tender offers in the sample.
Figure 3.1 plots a time-series of number of deals for the full sample, the U.K. domestic sample, and
the cross-border sample. Overall, the sample period analysed covered the fifth and the sixth merger
waves, from 1993 to 1999, and started from 2003, respectively. The number of deals for the three
samples share a similar trend over the sample period. The number of deals increased throughout the
1990s, and peaked in 1999. They started to increase from 2003 after a sharp decrease in 2000, the
year the stock market crashed. The number of deals increased for several years from 2003 and 2006
before they declined in 2007.
Figure 3.2 plots the time-series of total deal value for the full sample, the U.K. domestic sample, and
the cross-border sample. The overall trend of the total deal value is consistent with that reported in
Figure 3.1. Interestingly, the overall trend of the total deal value of the cross-border acquisitions has
changed more dramatically than that of the domestic deals, and is larger than that of the domestic
sample in certain periods, such as 1999 to 2000, 2005 to 2007, and 2012 to 2013.
Table 3.2 shows the distribution of bidder nations. The sample consists of bidders from 13 countries.
Amongst them, a large majority are U.K. bidders, taking up 74.42% of the full sample. Of 451 cross-
border acquisitions, 67 acquisitions involve U.S. bidders who dominate foreign bidders in the cross-
border acquisition sample, taking up 43.22%. These results may imply that U.S. firms who share the
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
110
same language and have a similar cultural background to the U.K. firms are more likely to undertake
an acquisition of a U.K. target.
The thesis defines the offer premium (the target 52-week high) as the logarithmic term difference
between the offer price (the target’s highest stock price over 335 calendar days ending 30 days prior
to the announcement date) and target stock price 30 days prior to the announcement date.35 The
measure of offer premiums reflects the target expected gains during M&As, a concept which has been
widely used in related literature (Betton et al., 2009, Eckbo, 2009).36
A histogram of the difference between the offer premium and the target 52-week high, as shown in
Figure 3.3. The x-axis is in the range of -500% to 500%, and the y-axis shows the density. The shape
of the histogram indicates the extent to which the target 52-week high approaches the offer premium
within the sample.
The reference point effect on offer premiums was studied by controlling for a series of deal, bidder,
and target characteristics, as reported in Table 3.3. Edmister and Walkling (1985) documented that
cash payments are associated with higher offer premiums compared with stock payments, since
targets require higher offer premiums to compensate for the tax expense generated by capital gains.
Faccio and Masulis (2005) suggest that diversified mergers lead to greater uncertainty about bidder
35 Offer Premiums = log (Offer Pricei,t) – log (Stock Pricei,t-30). Target 52-week high = log (Target 52-week Highest Stock
Pricei,t-30) – log (Stock Pricei,t-30). The next trading day’s stock price was used when an offer was announced at a weekend.
The logarithmic term was used to counter the positive skewness bias of offer prices.
36 Prior research measures offer premiums with target abnormal returns (Schwert, 1996) questioned by Eckbo (2009). The
results are robust when a wide range of measures used for offer premiums, including target abnormal returns similar to
Schwert’s approach and the target’s price in a week and 3 months before the announcement date.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
111
value, which implies that targets should be reluctant to accept risky offers unless higher offer
premiums are offered in compensation. A positive relationship between offer premiums and hostile
acquisitions is expected, since bidder managers pay higher offer premiums to persuade target
managers to accept an offer that may risk their professional career following a takeover. Bradley et
al. (1988) suggest that tender offers are associated with higher offer premiums since payment for the
manager goes to the shareholder.
Relative size is the deal value divided by the bidder market value. The market value (MV) is defined
as the current share price multiplied by the number of ordinary shares, expressed in the logarithmic
form. The market-to-book value (MTBV) is defined as the market value of the ordinary (common)
equity divided by the balance sheet of ordinary (common) equity in the company. Target volatility is
the standard deviation of target daily returns over the 335 calendar days ending 30 days prior to the
announcement. Run-ups are the pre-bid run-up prices calculated from 365 calendar days prior to the
takeover announcement date to seven calendar days before the takeover announcement date [-365, -
7]. All continuous data were winsorised at 1% and 99% levels to eliminate the outlier effect that both
extremely small or larger figures bias our results.
Relative size is defined as deal value divided by bidder’s market value, which measures the deal scale.
Previous research has highlighted a significant impact of the size effect on the offer premium (Asquith
et al., 1983; Dong et al., 2006). Firm size was measured as logarithmic term of market value of firms.
The mean value for firm size in the sample is larger for bidders than targets, 6.588 to 4.894, suggesting
that takeover bidders are generally stronger than their targets. This result is consistent with prior
M&A literature (Fuller et al., 2002; Moeller et al., 2004). A positive relationship between the offer
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
112
premium and bidder size, and a negative relationship between the offer premium and target size are
expected. According to Moeller et al. (2004), larger bidders tend to pay more for targets as they are
concerned with fewer restrictions in utilising the firm’s resources and thus become overconfident.
Alexandridis et al. (2013) suggest a robust negative relationship between the offer premium and target
size, implying that bidders’ acquisitions of a larger target are followed by a more complex process
for synergy creation, leading bidders to pay for targets with lower offer premiums in exchange for
expected synergies.
Firm’s growth opportunities were measured with market-to-book value (MTBV), which is consistent
with Rau and Vermaelen (1998). A higher MTBV indicates that the firm has a better investment
opportunity, whilst a firm with a lower MTBV suggests that the firm is short of investment
opportunities. Mean (median) bidder market-to-book value for the sample is on average higher than
that of the target, which suggests acquisitions involve a higher growth opportunities bidder and a
lower growth opportunities target. This is also predicted by the Q hypothesis of M&As (Lang, Stulz,
and Walkling 1989). A higher Q bidder is more likely to create synergies through acquiring a lower
Q target, thus it is willing to pay higher offer premiums for the targets. It is expected that offer
premiums are positively correlated to the bidder MTBV and are negatively correlated to the target
MTBV. A similar prediction can be made when the MTBV is a proxy for the firm’s valuation as per
Dong et al.’s work (2006). Thus, a high MTBV firm indicates that the firm is likely to be perceived
as a more overvalued firm whilst a low MTBV firm suggests the firm is likely to be perceived as
undervalued. It should be expected that a higher MTBV bidder in an attempt to dilute their
overvaluation tend to offer higher offer premiums to acquire a firm that is potentially being
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
113
undervalued (i.e. the lower MTBV targets). Therefore, in both cases, the offer premium is positively
(negatively) related to the bidder (the target) MTBV.
Pre-takeover price run-ups of both the bidder and the target (RunUps) were also controlled for while
analysing the reference point effect on the offer premium. Schwert (1996) found that the target price
run-ups increase the takeover costs of the bidder, indicating a dollar increases in target price run-up
leads to an increase of 1.13 dollars for the offer premium. It is suggested that the price run-ups are a
result of the market-wide valuation, leading the researcher to control for both the bidder and target
price run-ups. A mean (median) price run-up for the bidder is higher than that for the target. In
addition, the chapter accounts for the target volatility (Volatility) using the standard deviation of
target daily returns for the 335 calendar days ending 30 days prior to the announcement date.
Volatility of the firm represents the information asymmetry of the firm. A firm with high information
asymmetry tends to signal to the market that the firm is associated with high risks. Therefore, it can
be expected that the bidder tends to pay a higher offer premium when the true position of the target
firm is hard to justify.
The reference point effect on bidder announcement returns has been investigated by controlling for
a set of deal and bidder characteristics, since these factors have significant impacts on bidder
announcement returns, which are documented in prior M&A literature (Travlos, 1987; Rau and
Vermaelen, 1998; Alexandridis et al., 2010). More specifically, the method of payment for
acquisition (i.e. whether a means of payment for finance an acquisition is purely financed by stocks,
or cash), deal relatedness (i.e. whether or not the two firms involved in an acquisition is in the same
industry), deal type (i.e. tender offer or merger) and deal attitude (i.e. whether the deal is unsolicited)
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
114
were also accounted for. As to bidder characteristics, bidder size, bidder growth opportunities and
bidder pre-takeover period price run-ups were also taken into consideration.
Table 3.4 reports the mean value of a list of variables for two samples studied in this chapter: the
cross-border and the domestic acquisitions. The mean offer premium paid by the cross-border bidder
is 34.7%, which is significantly higher than that paid by the domestic bidder (27.6%), with a mean
difference of 7.1% at 1% significance level. There is no significantly difference for bidder abnormal
returns calculated by either the market-adjusted model or the market model. The mean value for the
target 52-week high is 27.7% for the cross-border acquisition sample, which is 4.4% higher than for
domestic acquisitions, suggesting that the reference point effect is more pronounced in the cross-
border acquisition sample than in the domestic acquisition sample. The cross-border bidder is found
to be stronger than the domestic bidder, reflected in significantly larger MV and higher MTBV than
those of the cross-border acquisition sample, suggesting that cross-border bidders who are more
disadvantageous in terms of information than domestic bidders should be strong enough to overcome
information asymmetry. The statistics further indicate that the information asymmetry for the cross-
border bidders tends to be larger than for their domestic counterparts.
3.4.2. Methodology
3.4.2.1. Piecewise linear regression and local polynomial smooth procedure
A non-linear relation between the offer premium and the target 52-week high is expected, based on
the shape of the value function proposed by prospect theory. According to prospect theory, targets
have a reference point in mind when selling the firms. Target shareholders may feel loss about the
deal their managers make when they receive a price that is lower than their 52-week high. This loss-
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
115
aversion occurs when targets lose the control of the firm. They may argue that the firm can perform
at a level similar to their 52-week high if their managers had not accepted the offer. Therefore, bidders
should take this into consideration and offer a price that is similar to the target 52-week high to
compensate the target’s mental feeling about the loss of control of the firm.
The OLS estimator was not applied in this context as it is a linear model assuming that the offer
premiums consistently increase with the target 52-week high. However, it is not the case as bidders
may walk away if the target’s current price is hard to capture its 52-week high price. Therefore, the
piecewise linear regression was employed to examine the relationship between the offer premium and
the target 52-week high. The piecewise linear function is a function of different linear segments, and
does not require an assumption of the shape of the data, which facilitates the analysis. It is worth
noting that bidders may not be able to offer a price to the target, whose current price is unlikely to
reflect their 52-week high. Bidders who pay according to such a 52-week high may find it hard to
rationalise the M&A motive to their shareholders. In addition, the marginal pain of the target
decreases when mental loss increases, leading additional costs according to the target reference point
to be reductant. Therefore, bidders may push up a price to acquire a target with a price closer to the
reference price and may not be willing to pay higher offer premiums for a target with a significantly
lower current price than the reference price. This suggests a consideration of diminishing sensitivities,
which is one implication of the prospect theory, while building the model to study the reference point
effect on the offer premium. The value function suggests that in both loss and gain domains, marginal
considerations tend to diminish relative to a reference point.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
116
Though the quadratic function has been widely used in estimating the non-linear relationship, Gould
(1993) argue that the quadratic estimator has a reverse effect, which makes the slope hard to be
determined. The author suggests that the piecewise linear function outperforms both the linear
function and quadratic function in fitting the true shape of observations. Morck et al. (1988)
conducted a piecewise linear function to examine the relationship between management ownership
and the firm’s market value.
The local polynomial estimator will be employed, such a technique is to take the average of several
neighbourhood points and smooths the chart. Avery (2013) suggests that a local polynomial estimator
is the best linear smoother for point estimation. As depicted in Figure 3.4, the offer premium increases
with the target 52-week high, and this relationship can be observed clearly in three linear segments
with different slopes. The lower linear segment lies below 30% of the target 52-week high, the middle
linear segment is in the range of 30% and 70% of the target 52-week high, and the upper linear
segment is above 70% of the target 52-week high. 37 The piecewise linear function is presented in the
following, where the coefficient is the change in slope from the preceding group.
, 30(1) min 52 ,0.3i tf weekhigh (3.1)
, 30(2) max 0,min 52 0.3,0.4i tf weekhigh (3.2)
, 30(3) max 52 0.7,0i tf weekhigh (3.3)
37 The F-test and the Chow test were performed to examine the break points for the three target 52-week high segments
and the results show that the three segments are significantly different.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
117
, 1 2 3 ,(1) (2) (3)i t i tOfferPremiums f f f (3.4)
where , 3052 i tweekhigh is defined as the logarithmic term difference between the target’s highest
stock price over 335 calendar days ending 30 days prior to the announcement date and the target stock
price 30 days prior to the announcement date. f (1) through f (3) are the transformation of the target
52-week high, the sum of f (1) through f (3) is equal to the target 52-week high. 1 ,
2 and 3 are
the coefficients for each linear segment, which represent the slopes of each segment.
3.4.2.2. Short-term event study
Short-term event study is a statistical tool for empirical research, and captures the market reaction
around the announcement date of M&As. Franks and Harris (1989) suggest that changes in stock
market should be fully captured in the several days around takeover announcement dates. Following
Brown and Warner (1985) who suggest that the standard procedures are well-specified and have
indicative power on firm performance, the daily stock returns with the short-term event study is
examined.
3.4.2.2.1. Market-adjusted model
Brown and Warner's standard short-term event study (1985) was used to calculate the CARs for the
bidder and the target in the five-day window (-2, 2) around the announcement date. The share returns
from the daily share price were calculated as follows:
, , , 1ln lni t i t i tR P P (3.5)
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
118
where ,i tR is the daily normal return of the firm i on day t, ,i tP denotes the share price of firm i on
day t, and , 1i tP denotes the firm’s share price lagged by a day.
, , , 1ln lnm t m t m tR P P (3.6)
where ,m tR relates to the daily normal return of the firm i while ,m tP and , 1m tP refer to the stock price
of the firm on day t and lagged by a day, t-1. The FTSE All-Share Index was used to proxy for the
value-weighted market index in the United Kingdom. The total market country index was used for
the other countries, which is defined as “TOTMK” plus “Country Code” in Datastream. For example,
US market index is “TOTMKUS”. The difference in the logarithm of the price was used performed
to capture the compound effect of the share price (Fama et al., 1969, Brown and Warner, 1985). The
abnormal returns was calculated with the market-adjusted model:
, , ,i t i t m tAR R R (3.7)
where ,i tR denotes the return of firm i on day t and ,m tR denotes the value-weighted market index
return of firm i on day t. The market-adjusted model assumes that α = 0 and β = 1 in our sample. The
abnormal returns (i.e. ARs) are summed to yield the CARs for the firm over the five-day window (-
2, 2) around the announcement date as follows:
, ,
0
n
i t i t
i
CAR AR
(3.8)
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
119
The market-adjusted model, as one of the main standard event methodologies, is widely employed to
measure firm takeover performance around the announcement date (Dong et al., 2006, Guo and
Petmezas, 2012). The model is believed to be particularly appropriate to assess the ARs of the firm,
as the model takes the market wide influence on the individual firm into consideration. We use
market-adjusted model to calculate CARs three and five days around the takeover announcement
date.38 The 5-day CARs are reported as our main results while the 3-day CARs are used as a
robustness test.
In order to assess the mean difference of CARs for the subsamples of domestic and cross-border
M&A deals, we employ t-statistics which is according to Seiler (2004). It allows the researcher to
investigate whether there are any distinguished characteristics of each portfolio CARs. The formula
is presented as follows:
( ) /
T
T
ARt
AR n (3.9)
where TAR denotes to the sample mean and as noted by Lyon et al. (1999:173), ( )TAR denotes to
the cross-sectional sample standard deviation for the sample of n firms.
3.4.2.2.2. Market Model
M&A literature also suggests abundant methods to be available to detect the short-term abnormal
returns. The market model was used as an alternative short-term event method to calculate the firms’
38 See also Fuller et al. (2002) and Faccio and Masulis (2005).
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
120
announcement returns, which is following Eckbo et al. (2016), who used the market model for a U.S.
study, and Gregory (1997), who used this method for a U.K. study. The model is specified as follow:
1it mt itR R (3.10)
where itR denotes holding period returns for firm i in the period t,
mtR denotes the value-weighted
market index. Similar to the market-adjust model, the FTSE All-Share Index was employed as a proxy
for the U.K. and total market country index was used for the other countries, which is defined as
“TOTMK” plus “Country Code” in DataStream, as described in the methodology session of market-
adjusted model. it denotes the error term. The market model parameters were estimated over the
window from 261 to 28 trading days prior to the announcement date [-261,-28], and calculate CARs
in both 3-day and 5-day event windows, and reported as robustness tests.
3.4.2.3. Multivariate analysis
The multivariate regression was used to examine the effect of offer premiums and the reference point
on firms’ short-run performance. According to Draper and Paudyal (2008), multivariate analysis is
superior in analysing the causation relation between the market reaction and related variables.
, , ,
1
N
i t i i t i t
i
CAR X
(3.11)
where ,i tCAR denotes the short-run CAR of firm i on day t, ,
1
N
i i t
i
X
denotes all the explanatory
variables in the model, and ,i t denotes the error term.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
121
3.4.2.4. Two Stage Least Squares (2SLS)
A 2SLS estimator was constructed to test the effect of offer premiums on bidder performance around
the announcement date using the target 52-week high as an instrumental variable (IV). The use of IV
is to control endogenous variable. In this chapter, the regression of the offer premium on bidders
CARs was suspected with endogeneity issues. If the market views an offer price based on the
reference point as an overpayment, there are more negative market reactions generated by the 2SLS
estimator than the OLS estimator, and if the market views an offer price based on the reference point
as synergies, there are more positive market reactions generated the 2SLS estimator than the OLS
estimator. With the target 52-week high, the chain response of the offer premium to the bidder
announcement returns was able to examine. The use of the target 52-week high as an instrumental
variable is based on the theory of the offer premium. Larcker and Rusticus (2010), in their study
regarding the use of instrumental variables, suggest that instrument variables are eventually justified
by the theory. Equations of the two stages of the 2SLS estimator were presented as follows:
First stage:
(3.12)
Second Stage:
, 1 , , ,
1
N
i t i t i i t i t
i
CAR OfferPremiums x
(3.13)
where ,i tOfferPremiums in the first stage regression were calculated as the logarithmic term
difference between the offer price and target stock price 30 days prior to the announcement date.
,i tOfferPremiums , presented in the second stage, were the fitted value obtained from the first stage
, 1 , ,52i t i t i tOfferPremiums weekhigh
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
122
regression. ,52 i tweekhigh is the target reference point, which is defined as the logarithmic term
difference between the target’s highest stock price over 335 calendar days ending 30 days prior to the
announcement date and target’s stock price 30 days prior to the announcement date. The fitted value
from the first stage regression was obtained and used to replace the offer premiums in the second
stage regression. The target 52-week high is used as an instrumental variable of offer premiums as it
reflects the target performance, which is uncorrelated with bidder announcement returns and strongly
correlated with the offer premium.
The validity tests of the use of IV by performing Hausman test as well as the first-stage F-test were
conducted. The null hypothesis of Hausman test is that OLS has no measurement error or OLS
estimates are efficiency. The null hypothesis of the first-stage F-test suggests that there is no relation
between the endogenous variable and the instrumental variable (or there is no weak instrument
problem). According to Stock et al. (2002), the sampling distribution is generally nonnormal when
weak instrument problems exists, which lead to an unreliable estimation. Therefore, the first-stage F
test was employed to combat this problem, Stock et al. indicate the first-stage F-statistics should be
large enough to reject the null hypothesis.
3.5. Empirical results
3.5.1. The reference point effect on the offer premium
In this session, the piecewise linear regression of the offer premium on the reference point effect will
be conducted. Based on the prediction of prospect theory value function, our hypothesis proposes that
there is a positive correlation between offer premiums and the reference price in the U.K. market.
The main variable of interest to be investigated is the target 52-week high. Other variables that have
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
123
important impacts on the offer premium according to the M&A literature were controlled for in the
regressions.
The researcher first examines the correlation between the offer premium and the target 52-week high
with kernel density graph. As reported in Figure 3.3, the offer premium is highly correlated to the
target 52-week high. In most cases, offer premiums are paid around the target 52-week high,
suggesting that the reference point effect is pronounced in the sample.
Based on the diminishing sensitivity, it was expected that the offer premium tends to be more relevant
to the target 52-week high in the lower 52-week high level for the rationale that the target’s current
price is likely to reflect its recent best performance, leading the bidder to believe that it is possible to
generate synergies following an M&A deal, while the offer premium is less relevant to the target 52-
week high in the higher 52-week high level, implying that the bidder may walk away when the target’s
current price is unlikely to reflect its recent best performance.
Figure 3.4 supports our prediction by portraying a non-linear relationship between the target 52-week
high and offer premiums. The function predicts that targets have less marginal consideration about
loss. As noted earlier in the methodology section, the first linear segment lies below 30% of the target
52-week high (i.e. Low52-week high), the second linear segment lies between 30% and 70% of the
target 52-week high (i.e. Mid52-week high), the third linear segment lies above 70% of the target 52-
week high (i.e. High52-week high).
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
124
Table 3.5 shows a positive relationship between the offer premium and the target 52-week high, which
supports our hypothesis H1a. Specification (1) shows that a 10% increase of the target 52-week high
leads to an increase of 2.08% of the offer premium (coefficient = 0.208, t = 5.180). As we expected
earlier, the offer premium is less relevant to the target 52-week high when the level of the target 52-
week high is high or the current price is unrealistically far below the target 52-week high. Thus, it
leads us to study the reference point effect on the offer premium in the three different segments
predicted in the piecewise linear regression: the low target 52-week high, the mid target 52-week high
and the high target 52-week high. Specification (2) reports the target 52-week high on the offer
premium without any other control variables. It can be seen that the fitness of the specification is only
5.5%, which means the model might not be well-specified. Three different sets of variables in the
specifications (3) to (5) were taken into accounted. Specifically, the deal and target characteristics
were controlled in specification (3), deal and bidder characteristics were controlled for in
specification (4).
Specification (5) of this table shows the main results for the reference point effect on the offer
premiums. The offer premium is less relevant to the target 52-week high when the level of the target
52-week high increases. The offer premium increases by 4.33% for a 10% increase in the low level
of the target 52-week high, and increases by a 2.05% for a 10% increase in the middle level of the
target 52-week high, whereas the high level of the target 52-week high is insignificantly related to
the offer premium.39 Results obtained in this table suggest that bidders pay according to the target 52-
week high. However, they are rationally limited when the target 52-week high is unable to provide
any rationale motive for a higher price. In that case, bidders may find it hard to pay according to the
39 Our results are also consistent when controlling for the firm’s leverage and year-, industry-, and firm-fixed effects.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
125
target 52-week high, which could create wealth for their shareholders when the target’s current price
deviates unrealistically far from the target 52-week high. The results obtained in this table can be also
interpreted within the prospect theory framework: while the price that deviates more from the
reference point tends to decrease the marginal considerations about the loss of the target (if the target
views selling the firm as a loss), leading higher M&A offer premiums are reductant for the bidder.
By doing so, bidders tend to be reluctant to pay offer premiums according to the target 52-week high
when the target 52-week high is high. The results are more pronounced compared with the U.S.
findings of Baker et al. (2012).
The results reconcile several predictions of prospect theory. People have reference-dependence bias
in that their decision on the offer premium is based on a salient piece of price information. Specifically,
the offer price is found to be set based on the target 52-week high. Bidders use the target’s recent best
performance as a reference price to suggest that the primary M&A motive for the bidder is synergy
generation, since targets whose reference point price is closer to their current performance is likely
to convince bidders to pay according to their reference prices.
The results are also in line with the diminishing sensitivity of prospect theory. It can be seen that the
lower level of the target 52-week high has more substantial impacts on the offer premium than the
higher levels. Bidder managers in the lower segment of the target 52-week high tend to believe
acquiring targets whose current price deviates greatly from the reference price is likely to leave more
room for synergy exploration and thus pay generous for targets than the scenario when the target’s
current price is closer to the reference price. While in the higher segments of the target 52-week high,
bidder managers tend to believe the target 52-week high is unable to deliver real support for the true
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
126
performance of the target, leaving the reference point effect on the offer premium less relevant. This
is in line Baker et al's view (2012) that bidders are cautious about those targets, whose current price
deviates greatly from the 52-week high, since they are more likely associated with bankruptcy risks,
violating the reference point effect. It is also possible that target managers are unable to bargain a
higher offer premium based on a high level of reference points.
Further, investors generally have a loss-aversion tendency, and such losses are gauged in relation to
the reference point. Based on these results, bidders pay a price related to the target 52-week high to
compensate for target shareholders’ mental loss when giving up the control of the firm. Target
shareholders, whose firm’s current price is close to the 52-week high, may presumably believe that
the firm would be better off if it was not acquired. In that case, bidders have to compensate the targets’
loss with an offer premium.
In addition to those explanations above in relation to prospect theory, it is expected that the stronger
legal environment and shareholder protection in the U.K. market than any other market may result in
a stronger reference point effect. One possible explanation is that in a market where the interests of
target managers and shareholders are closely aligned, managers are less likely to conduct any
frustrated actions against their shareholders. Thus, target managers are likely to reinforce their
bargaining power in M&As.40
40 The effect of corporate governance on the firm’s bargaining power is a recommendation for future research highlighted
in the Chapter 6 of this thesis.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
127
It is worth noting that that the variable Cross Border is positively related to the offer premium,
reported in Table 3.5. Cross Border is a binary variable, taking value of 1 if acquisitions are
undertaken by cross-border bidders and 0 otherwise. Our results are in accord with the view of Rossi
and Volpin (2004) that stronger shareholder protection results in lower costs of capital favouring the
intense competition environment, increasing offer premiums. According to Edmister and Walkling
(1985), the offer premium represents the bargaining power of negotiators. In the U.K. settlement,
cross-border bidders have less bargaining power than domestic bidders engaging in acquisitions,
resulting in larger M&A offer premiums.
According to this prediction, the reference point effect will be further tested by cross-border
acquisition and domestic acquisition subsamples. It is proposed that there is a positive correlation
between offer premiums and the reference price when cross-border bidders acquire U.K. targets.
Targets with greater bargaining power are more likely to use the reference point to require a higher
offer premium.
Table 3.6 shows a slightly stronger reference point effect on the offer premium for the cross-border
acquisition subsample than for the domestic acquisition subsample. Several possible explanations for
this finding are offered. First, it is suggested that geographical difference creates greater information
barriers for cross-border bidders than for domestic bidders, which weakens cross-border bidders’
bargaining position. With less information about the target, cross-border bidders are likely to estimate
the target value with more relevant and easily obtainable information such as the target 52-week high.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
128
Further, it can be argued that one of the main takeover motives for cross-border bidders is to access
the foreign capital market to integrate their resources, and obtain strong shareholder protection when
bidders from a country with relatively weaker shareholder protection than the target country. With
this in mind, cross-border bidders are likely to be influenced by the target 52-week high. According
to Alexandridis et al. (2013), acquiring large targets is a complex task which requires considerable
information, while it is impossible to assess the firm value given limited time and information. Thus,
valuing targets based on its target 52-week high simplify the acquisition process, giving cross-border
bidders quick access to the U.K. market. It is quite common that cross-border bidders are relatively
those large companies, as they should be strong enough to overcome barriers. According to Moeller
et al. (2004), managers from large companies have easier access to the firm’s resources, making them
more likely to overpay for targets.
Information disadvantages arising from geography weaken bidder bargaining position. Compared
with Baker et al.’s work (2012) focusing on the reference point effect on U.S. domestic M&A deals,
results obtained in this chapter show that the target 52-week high has greater predictability for offer
premiums for both cross-border and domestic acquisitions in the United Kingdom. Cross-border and
domestic bidders tend to have similar bargaining power, as reflected in the coefficients of the target
52-week high. However, the size effect distinguishes the two subsamples. Bidder size is more
significant for domestic acquisitions than for cross-border acquisitions, suggesting that domestic
bidders tend to be more consistent with the hubris hypothesis cited in Moeller et al.’s work (2004),
affirming that a larger firm is likely to be more overconfident than that of a smaller firm. The further
analysis aims to explore why cross-border bidders tend to pay similar M&A premiums relative to the
target 52-week high as domestic bidders.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
129
In Table 3.6, the offer premium is positively related to the relative size, bidder size, the pre-bid target
run-up price and hostile mergers. It is also found that the offer premium is negatively related to target
size, target MTBV, target stock volatility, all-stock financed acquisitions and all-cash financed
acquisitions. These results are consistent with prior research (Schwert, 2000; Betton et al., 2008;
Eckbo, 2009). It is worth noting that the size of the bidder and target play an important role in offer
premiums. A larger bidder tends to pay generously for a smaller target, consistent with the findings
of Moeller et al. (2004) and Alexandridis et al. (2013). Contrary to the tax implication hypothesis, it
is suggested that the offer premium is negatively related to all-cash acquisitions, as reported in
specification (4). It can be interpreted as follows: cash is the greatest liquidity asset that can instantly
meet the target’s demands, thus the use of cash increases the bidder’s bargaining power, reducing
offer premiums.
To ensure that the model used here does not suffer from a muliticollinearity problem, i.e. the main
variable has enough explanatory power to the dependent variable, a test of variance inflation factor
(VIF) is employed. The test assumes that there is no correlation between the regressors when the VIF
value falls below 10. The presence of a multicollinearility problem in the model though does not
reduce the predictability of the model as a whole, however it increases the variance of the model,
making the model unstable, and reduces the accuracy of each individual variable included in the
regressions. The results obtained show little evidence of multicollinearity problems.41
41 It was found that the mean value for VIF is 1.75, and there is no VIF value beyond 4. The highest VIF value is 3.32,
which is the bidder MV, the second highest VIF value is 2.94, which is the target MV. The chapter reports the main results
of VIF test but omits the table for the sake of brevity.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
130
3.5.2. The reference point effect on bidder announcement returns
This section examines the reference point effect on bidder announcement returns. A 2SLS regression
using the target 52-week high as an instrumental variable will be conducted. The rationale of using
the target 52-week high as an instrumental variable is because the offer premium is not a clean
measure, as it either represents the synergy or an overpayment. Hence, in order to discover the role
of the reference point effect on CARs, the 2SLS estimator is used. If the offer premium (driven by
the reference point effect) in the 2SLS leads to fewer announcement returns for the bidder than in the
OLS, the reference point plays a role of overpayment, otherwise, the reference point indicates the role
of synergies. It is proposed that bidder short-run performance is negatively correlated to offer
premiums when deals are made with regard to the reference price.
Table 3.7 shows that bidder announcement returns are either insignificantly or significantly
negatively related to the offer premium in different subsamples, as reported in specifications (1), (3),
and (5) respectively. It appeared that cross-border bidders pay higher offer premiums but outperform
domestic bidders. An acquisition of a U.K. target is good news to shareholders of cross-border bidders
who might believe that their rights to be protected in a more stringent legal environment, and thus
react positively to the bid announcement.
It is expected that the target 52-week high increases the offer premium, which further decreases the
bidder announcement returns. The Hausman test shows a p-value of 0.0004, suggesting that the 2SLS
estimator is more efficiency than the OLS estimator. The first-stage F-test shows a p-value of 0.000042,
indicating that the target 52-week high is a valid instrumental variable for the offer premium. A more
42 See Stock et al. (2002) for a survey of weak instrument problems.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
131
negative relationship between bidder announcement returns and the offer premium became apparent
when using the target 52-week high as an instrumental variable compared with those without using
the target 52-week high as an instrumental variable, as shown in specifications (2) to (1), (4) to (3),
and (6) to (5) respectively. The market reacts more negatively to a bid announcement when bidders
pay based on the target reference price, suggesting that the reference price indicates an overpayment
in the U.K. M&As.43
These results are in line with prior M&A research. According to Rossi and Volpin (2004) and Moeller
and Schlingemann (2005), it is expected that the strength of shareholder protection in the U.K. would
weaken the bargaining power of cross-border bidders, thus using the target 52-week high is more
likely to receive shareholders’ consent. The sample was divided into cross-border and domestic
acquisition subsamples, as reported in specifications (2) and (3), and it was found that the target 52-
week high driven offer premiums were significantly negative related to announcement returns of
domestic bidders, and were not significantly related to those of cross-border bidders,44 suggesting
that shareholders of cross-border bidders acquiesce more easily with the target 52-week high as the
reference price. This chapter also provides additional support to the work of Kuipers et al. (2009),
who found that cross-border bidders in pursuit of stronger shareholder protection in the U.K. will pay
higher offer premiums as compensation. Further, the bidder size for cross-border acquisitions is more
negatively related to bidder announcement returns, which is consistent with the findings of Rau and
Vermaelen (1998) that small bidders tend to be more cautious than the large bidders. Taken together,
43 Similar conclusions can be drawn by using the market model in a window of 261 to 28 trading days prior to the takeover
announcement date. The results are reported in Table 3.9. 44 The obtained results of this chapter are robust to bidder announcement returns calculated with the market model.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
132
cross-border bidders who have less bargaining power might have to pay higher premiums according
to the target 52-week high.
3.6. Robustness checks
3.6.1. The target 52-week high as a reference point
In this subsection, robustness results are provided for the target 52-week high as a reference point by
first employing other price measures related to the target 52-weeks: the target 52-week minimum and
the target 52-week mean. Moreover, the target monthly returns prior to the takeover announcement
date are utilised the expectation that the bidder may also extrapolate the past performance of the target.
Results on regressions of the offer premium on the reference point are robustness after checking for
different price measures related to the target 52-weeks. The target 52-week minimum and the target
52-week mean were taken into consideration besides the target 52-week high in terms of the reference
point price, as reported in Table 3.8. Huddart et al. (2009) provided an overall review with regard to
the reference price using both 52-week highs and 52-week lows. In a similar way, the target 52-week
mean is also considered as the reference price. It is generally believed that the target 52-week prices
tend to attract a great deal of investors’ attention, making more conservative investors likely to focus
on the firm’s mean value over 52 weeks prior to the announcement date since it reflects the firm’s
performance in a certain period of time rather than a snapshot price. The results show that the target
52-week high remains strong after checking for those proposed reference point price measures.
In addition, robustness checks were carried out by including the target past returns prior to the
takeover announcement date when studying the reference point effect on the offer premium (see Table
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
133
3.9). According to Rau and Vermaelen’s extrapolation hypothesis (1998), market investors tend to
forecast a firm’s prospects based on its past returns, creating a strong performance momentum around
a bid announcement for those with best past performance. Therefore, targets’ past returns of each
month were taken into consideration, reflecting the firm’s returns over the past months, which are
more likely be a proxy for the market perception of the firm’s valuation than a single item of price
information (i.e. the target 52-week high). The extrapolation effect on the offer premium was studied
by including the target past returns up to 12 months prior to the takeover announcement date. In
particular, specification (2) controls for the target monthly returns for the recent three months prior
to the takeover announcement date, specification (3) controls for the target monthly returns for the
recent six months prior to the takeover announcement date, and specification (4) controls for the
target monthly returns up to a year prior to the takeover announcement date. It was found that the
offer premium is positively related to the target 52-week high. Once again, our results show that the
target 52-week high is a proper measure for the reference point in an M&A deal.
3.6.2. The reference point effect on the offer premium by market conditions
In addition, the reference point effect on the offer premium by the market wide valuation was
examined. Investors’ have a rationale that bidders may under- or over-react to a target’s valuation
when the market valuation fluctuates. The sample periods are classified into High-, Low- and Neutral-
valuation markets according to Bouwman et al.'s method (2009). Results of Table 3.10 show that the
reference point effect is strong when the market-wide valuation is either high or low. It can be
interpreted that a high market valuation is likely to push the target’s current price closer to the 52-
week high, making bidders to believe that they can outperform the target’s recent performance. While
a low market valuation makes the bidder to believe that, the target 52-week high is a true indicator
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
134
for the target’s performance given that no other factors drive the firm’s value beyond or below its
fundamental value. Moreover, it was found that cross-border acquisitions tend to pay higher offer
premiums when the market valuation is high, implying that the high market wide valuation increases
information asymmetry and thus reinforces the target’s bargaining position over cross-border bidders.
3.6.3. The reference point effect on the offer premium across different subsamples
The reference point effect on the offer premium was studied by different nations with the expectation
that bidders whose nation was farthest away compared with that of the U.K. firms would tend to have
greater information barriers and be more likely to rely on the target 52-week high. In addition, the
reference point effect on the offer premiums in different M&A subsamples was analysed: the payment
method for the M&A deals, whether or not the deal is diversified and whether the deal involves a
negotiation between the boards of the two firms (i.e. tender offers and mergers).
A target’s bargaining position is likely to be reinforced when the distance involved between the two
firms’ home countries is large. The M&A sample was divided into two subsamples, the one includes
acquisitions comprising U.K. bidders, the other comprising acquisitions involving U.S. bidders,
farthest away compared with other cross-border bidders (see Table 3.11). It can be seen that the offer
premium increases by approximately 6% in the U.S. bidders compared with 4% in the U.K. domestic
bidders. Greater distance between the two nations impedes the transaction, making the bidder more
likely to look at the target reference point. The results obtained in this table are consistent with those
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
135
of Erel et al. (2012) and Uysal et al. (2008), finding that geographical distance plays a role in
information dissemination.45
The reference point effect on offer premiums across different subsamples by deal information was
also tested. The reference point effect on the offer premium being tested by the method of payment
(see Table 3.12), by whether or not the deal is diversified reported in (see Table 3.13), and the type
of merger reported in (see Table 3.14). Taken together, the results show that the target 52-week high
as a proxy for the reference point effect is very pronounced in different subsamples, and has a
substantial impact on the offer premium in different M&A subsamples.
3.6.4. The reference point effect on bidder announcement returns
Table 3.15 tabulates results on regressions of the offer premium on the bidder announcement returns
and with the target 52-week high as an instrumental variable of the offer premium. Instead of using
the market-adjust model to calculate the firm’s abnormal returns around the announcement date, the
market model is also used. The market model parameters were estimated based on an event window
from 261 to 28 trading days prior to the announcement date, i.e. [-261,-28]. The results obtained from
the market model are consistent with those by the market-adjusted model. More specifically, a 10%
increase in the reference point driven offer premium decreased the bidder announcement returns by
1.83% in the full M&A sample. While dividing it into two subsamples, the reference point effect leads
to more negative bidder announcement returns for domestic bidders than for cross-border bidders.
45 It can be argued that the information asymmetry are relatively low due to culture similarity between the two countries.
However, the sample included in this chapter is small, which is likely to result in bias in the results. Further investigation
by including more observations needs to be done to explain whether the culture or the distance between the countries
matter the bargaining power. We leave it to the future research, since the focus of this chapter is to test the target reference
point effect on the firm’s valuation.
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
136
In addition, bidder abnormal returns in a 3-day window around the takeover announcement date were
examined using both market-adjusted model and market model. The reference point effect on bidder
3-day announcement returns calculated with the market-adjusted model is reported in Table 3.16, and
results with the market model are reported in Table 3.17, which are consistent with the main results
reported in Table 3.7.
3.7. Conclusion
In this chapter, the reference point theory of M&As in the U.K. context has been examined. It was
found that the reference price measured by the target 52-week high is significantly positively related
to the offer premium. The reference point effect plays an important role in valuing the target firm.
Both the bidder and the target have reference-dependence bias, but the rationales behind using the
reference point need to be clarified in terms of different parties. Targets with greater bargaining power
are likely to use the reference point to ask for more offer premiums. It can be argued that bidders
paying according to the reference point in favour of targets is either because of the overconfidence or
because of information barriers. However, it is evident that the reference point effect is no longer
sensitive to bidders when the target’s current price deviated greatly from its reference price (or the
high level of the target 52-week high).
Further, the sample was extended to cross-border acquisitions and it was found that the reference
point effect is more pronounced to the cross-border acquisition subsample than to the domestic
acquisition subsample. It can be interpreted as follows: targets have greater bargaining power over
cross-border bidders than over domestic bidders, and they are likely to translate it into higher offer
premiums. Cross-border bidders may find it difficult in overcoming information barriers and might
Chapter 3. Reference Point Theory on Cross-Border and Domestic M&A Deals: U.K. Evidence
137
have to pay an offer price based on the reference point. They might as well actively pay based on the
target 52-week high, the reason of being to reinforce their competitiveness to enter the United
Kingdom.
Finally, bidders paying based on the reference point receive negative market reactions, since
shareholders tend to believe that the managers are paying too much, thereby destroying their wealth.
The finding also suggests that bidder shareholders are pessimistic about their own profit-generating
ability. Together, results obtained in this chapter are consistent with the predictions of prospect theory.
Chapter 3. Tables & Figures
138
Table 3.1: Summary statistics for M&A sample
This table reports summary statistics for a sample of M&As announced between 1985 and 2014. The number N
is the number of deals per year. The third and fourth columns are the mean and the median of the deal value
respectively, reported in million U.S. dollars. The fifth and sixth columns present the method of payment
information. Number of stock only (number of cash only) is defined as the method of payment is 100% by stock
(cash). The number of diversified acquisitions, hostile acquisitions and tender offers are reported from the seventh
to ninth columns. “Diversification” refers to diversified deals in which the primary two Standard Industry
Classification codes are different between bidders and targets. “Hostile” refers to hostile bids. “Tender” refers to
tender offers.
Year N Deal Value ($mil)
Mean Median
No. of
Cash Only
No. of
Stock Only
No. of
Diversified
No. of
Hostile
No. of
Tender
1985 1 16.10 16.10 1 - 1 - -
1986 1 14.00 14.00 1 - 1 - -
1987 9 283.07 223.10 6 2 5 1 3
1988 10 135.52 72.36 8 - 5 - 4
1989 10 650.56 34.37 9 - 7 1 3
1990 3 225.56 115.07 2 - 1 - 1
1991 19 294.68 87.39 12 1 15 3 10
1992 8 248.12 31.89 4 2 7 1 4
1993 11 38.71 27.08 6 1 5 1 7
1994 21 160.01 42.18 15 3 8 2 10
1995 23 358.74 41.67 18 - 8 2 9
1996 20 644.40 133.14 10 6 14 1 11
1997 33 356.98 60.41 24 6 14 1 25
1998 56 655.03 78.95 34 14 21 1 40
1999 65 943.50 121.28 32 12 37 2 53
2000 58 533.48 153.60 37 12 28 3 47
2001 25 907.36 77.90 15 3 11 - 19
2002 20 946.02 24.98 14 4 3 - 19
2003 17 595.35 123.59 10 4 4 - 13
2004 22 461.28 120.23 12 7 13 - 16
2005 35 1037.71 330.44 20 7 19 2 25
2006 21 1082.69 210.80 12 4 10 2 19
2007 34 724.03 47.57 23 5 15 - 23
2008 13 376.26 55.36 10 1 7 1 11
2009 11 676.23 94.18 5 4 6 - 10
2010 26 315.46 22.95 19 3 15 - 9
2011 7 221.69 182.65 4 2 3 1 4
2012 13 444.14 64.08 10 2 4 - 6
2013 7 817.49 98.24 3 1 3 - 5
2014 7 570.73 83.14 4 2 2 - 2
Total 606 606.40 85.19 380 108 292 25 408
Chapter 3. Tables & Figures
139
Table 3.2: Distribution of bidder nations
This table presents the number of acquisitions and percentage of the number of acquisitions by bidder nations.
The summary statistic are based on a sample of 606 public acquisitions toward U.K. public targets announced
between 1985 and 2014. Bidders are publicly listed firms in their domestic stock markets.
Bidder Nation Number of acquisitions Percentage
Australia 5 0.83
Canada 9 1.49
France 13 2.15
Germany 16 2.64
Hong Kong 5 0.83
Ireland-Rep 4 0.66
Italy 5 0.83
Japan 9 1.49
Netherlands 7 1.16
South Africa 7 1.16
Switzerland 8 1.32
United Kingdom 451 74.42
United States 67 11.06
Total 606 100.00
Chapter 3. Tables & Figures
140
Table 3.3: Summary statistics for variables
This table presents the means, medians, and standard deviations of the variables. Panel A presents dependent
variables used in the regressions. Offer premiums are defined as the logarithmic term difference between the offer
price and the target stock price 30 days prior to the takeover announcement. Bidder 5-day CARs are bidder 5-day
announcement returns calculated by the market-adjusted model. Panel B presents independent variables, our main
variable the target 52-week high is defined as the logarithmic term difference between the target highest stock
price over 335 calendar days ending 30 days prior to the announcement and the target stock price 30 days prior to
the announcement. Deal characteristics, including the method of payment, deal attitude, type of deals, cross-border
deals and relative size. Here stock, cash, diversification, hostile, tender offer and cross-border are dummy variable,
taking value of 1 if acquisitions are 100% stocks, 100% cash, diversified merger, hostile merger, tender offer, and
cross-border acquisitions, and 0 otherwise. Relative size is the deal value divided by the bidder market value. The
market value (MV) is defined as the current share price multiplied by number of ordinary shares, expressed in
logarithmic form. The market-to-book value (MTBV) is defined as the market value of the ordinary (common)
equity divided by the balance sheet of ordinary (common) equity in the company. Target volatility is the standard
deviation of target daily returns over the 335 calendar days ending 30 days prior to the announcement. Run-ups
are the pre-bid run-up prices calculated from 365 calendar days prior to the takeover announcement date to seven
calendar days before takeover announcement date [-365, -7]. Accounting variables were collected one month prior
to the announcement date, and continues variables were winsorised at the 1% and 99% levels.
Panel A: Dependent Variables Mean Median Std Dev
Offer Premiums 0.294 0.263 0.206
Bidder 5-day CARs -0.002 -0.003 0.070
Panel B: Independent Variables
Target 52-week High 0.244 0.185 0.226
Deal Characteristics
Stock 0.178 - 0.383
Cash 0.627 - 0.484
Diversification 0.482 - 0.500
Hostile 0.041 - 0.199
Tender 0.673 - 0.469
Cross Border 0.256 - 0.437
Relative Size 0.353 0.150 0.549
Bidder Characteristics
Bidder ln(MV) 6.588 6.625 2.190
Bidder MTBV 2.490 1.760 4.617
Bidder RunUps 0.088 0.028 0.423
Target Characteristics
Target ln(MV) 4.894 4.760 1.846
Target MTBV 2.471 1.575 4.063
Target RunUps -0.013 0.005 0.382
Target Volatility 0.022 0.020 0.012
Number of Observations 606
Chapter 3. Tables & Figures
141
Table 3.4: Summary statistics for variables by cross-border and domestic subsamples
This table presents the means of variables by U.K. domestic and cross-border acquisition sample. The first column shows the mean value of variables in cross-border acquisitions.
The second column shows the mean value of variables in U.K. domestic acquisitions. Differential is the mean difference between the cross-border and the U.K. domestic. The
t-statistics are reported in parentheses and calculated using the t-test and the Wilcoxon rank-sum test for the difference between the cross-border sample and the U.K. domestic
sample. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
Cross-Border U.K. Differetial
(Cross-Border-U.K.)
Offer Premiums 0.347*** 0.276*** 0.071*** (19.492) (29.799) (3.743)
Bidder 5-day CARma -0.006 -0.001 -0.005 (-1.115) (-0.328) (-0.719)
Bidder 5-day CARmm -0.010* -0.003 -0.007 (-1.800) (-0.824) (-1.096)
Target 52-week High 0.277*** 0.233*** 0.044** (13.763) (22.783) (2.100)
Relative Size 0.211*** 0.401*** -0.190*** (8.531) (14.140) (-3.756)
Bidder ln(MV) 8.010*** 6.100*** 1.910*** (58.203) (60.994) (10.121)
Bidder MTBV 3.191*** 2.250*** 0.941** (13.000) (9.511) (2.197)
Bidder RunUps 0.089** 0.087*** 0.002 (2.474) (4.463) (0.048)
Target ln(MV) 5.538*** 4.673*** 0.865*** (39.692) (54.172) (5.135)
Target MTBV 2.761*** 2.372*** 0.389 (8.647) (12.307) (1.029)
Target RunUps 0.005 -0.020 0.025 (0.146) (-1.153) (0.696)
Target Volatility 0.025*** 0.022*** 0.003***
(24.188) (37.340) (2.883)
Number of Observations 155 451
Chapter 3. Tables & Figures
142
Table 3.5: The piecewise linear regressions of offer premiums on the target 52-week high
, 1 , 30 2 , 30 3 , 30 ,min 52 ,0.3 max 0,min 52 0.3,0.4 max 52 0.7,0i t i t i t i t i tOfferPremiums wh wh wh
This table reports piecewise linear regression results for offer premiums on 52WeekHigh, controlling for various deal and firm characteristics. We document three individual
target reference point (i.e. 52WeekHigh) levels: Low52wh (0-30%), Mid52wh (30%-70%), High52wh (70% or above). Colum (1) reports the relation between the offer premium
and the target 52-week high, column (2) reports the relation between offer premiums and the three individual levels about the 52WeekHigh, column (3) controls for target and
deal characteristics, column (4) controls for the bidder and deal characteristics, column (5) controls for bidder, target and deal characteristics. Variable definitions are as in the
notes of Table 3.3. Robustness t-statistics are reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively, is
reported alongside the coefficients.
(1) (2) (3) (4) (5)
Offer premiums Coef. t-stat. Coef. t-stat. Coef. t-stat. Coef. t-stat. Coef. t-stat.
52wh 0.208*** (5.180)
Low52wh 0.322*** (3.618) 0.440*** (4.800) 0.326*** (3.692) 0.433*** (4.912)
Mid52wh 0.122 (1.057) 0.221** (1.985) 0.139 (1.217) 0.205* (1.807)
High52wh 0.150 (0.451) 0.178 (0.549) 0.001 (0.002) 0.089 (0.260)
Stock -0.071*** (-2.933) -0.063** (-2.476) -0.079*** (-3.380)
Cash -0.077*** (-3.461) -0.052** (-2.283) -0.075*** (-3.522)
Diversification 0.010 (0.669) 0.018 (1.179) -0.005 (-0.342)
Hostile 0.080* (1.781) 0.047 (1.028) 0.084** (2.048)
Tender 0.065*** (3.509) 0.070*** (3.812) 0.028 (1.486)
Cross Border 0.085*** (4.258) 0.049** (2.256) 0.046** (2.294)
Relative Size -0.030* (-1.831) -0.023 (-1.318) 0.045** (2.375)
Target ln(MV) -0.023*** (-4.785) -0.060*** (-7.869)
Target MTBV -0.002 (-1.275) -0.004** (-2.417)
Target RunUps 0.112*** (3.638) 0.087*** (2.704)
Target Volatility -2.553*** (-3.641) -2.926*** (-4.238)
Bidder ln(MV) 0.006 (1.361) 0.045*** (6.337)
Bidder MTBV 0.001 (0.513) 0.002 (1.179)
Bidder RunUps 0.044** (2.013) 0.025 (1.187)
Constant 0.244*** (21.468) 0.229*** (16.271) 0.375*** (7.906) 0.164*** (3.430) 0.284*** (5.986)
N 606 606 606 606 606
R2 0.052 0.055 0.185 0.138 0.260
Chapter 3. Tables & Figures
143
Table 3.6: The piecewise linear regression of offer premiums on the target 52-week high of cross-border
and domestic acquisitions into the United Kingdom
This table reports the piecewise linear regression results for offer premiums on the target reference point of both
domestic and cross-border acquisitions into the U.K. market. It was documented three individual target reference
point levels: Low52wh (0-30%), Mid52wh (30%-70%), and High52wh (70% and above). Column (1) reports the
results for cross-border acquisitions. Column (2) reports the results for U.K. domestic acquisitions. Variable
definitions are as in the notes of Table 3.3. Equation is presented in the note of Table 3.5. Robustness t-statistics
are reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and
* respectively, is reported alongside the coefficients.
Offer Premiums (1) (2)
Cross-Border U.K. Domestic
Low52wk 0.480*** (2.779) 0.404*** (4.191)
Mid52wk 0.402 (1.600) 0.093 (0.768)
High52wk -0.149 (-0.226) 0.186 (0.469)
Stock -0.151** (-2.110) -0.072*** (-2.963)
Cash -0.063 (-1.241) -0.077*** (-3.285)
Diversification -0.010 (-0.310) -0.002 (-0.097)
Hostile -0.021 (-0.358) 0.157*** (2.748)
Tender 0.072* (1.931) 0.008 (0.377)
Relative Size 0.130* (1.682) 0.047** (2.451)
Target ln(MV) -0.044** (-2.202) -0.069*** (-8.700)
Target MTBV -0.004 (-1.402) -0.004** (-2.232)
Target RunUps 0.138** (2.518) 0.050 (1.370)
Target Volatility -3.049* (-1.967) -2.894*** (-3.576)
Bidder ln(MV) 0.024 (1.388) 0.054*** (7.266)
Bidder MTBV 0.015*** (2.651) 0.001 (0.666)
Bidder RunUps 0.023 (0.531) 0.024 (1.014)
Constant 0.317*** (2.701) 0.291*** (5.562)
N 155 451
R2 0.286 0.274
Chapter 3. Tables & Figures
144
Table 3.7: The OLS regression of bidder announcement returns on offer premiums
First stage:
Second Stage:
, 1 , , ,
1
N
i t i t i i t i t
i
CAR OfferPremiums x
This table presents both OLS and 2SLS regression results of bidder market-adjusted model 5-day announcement
returns (CAR5) on the offer premium. The target 52-week high is used as an instrumental variable of the offer
premium. We report the regressions for both cross-border and domestic acquisitions into the United Kingdom.
Hausman test and first-stage F-test are reported below the table. Variable definitions are as in the notes of Table
3.3. Details of the methodology were presented in the Methodology Section of this Chapter. Robustness t-statistics
are reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and
* respectively. (1) (2) (3) (4) (5) (6)
CAR5 Full Sample Cross Border U.K. Domestic
OLS IV OLS IV OLS IV
Offer Premiums -0.019 -0.207*** 0.015 -0.134 -0.036** -0.258*** (-1.360) (-3.268) (0.587) (-1.454) (-2.095) (-2.634)
Stock 0.007 -0.003 0.003 -0.012 0.008 -0.005 (0.807) (-0.291) (0.106) (-0.430) (0.762) (-0.408)
Cash 0.026*** 0.017* 0.001 -0.007 0.030*** 0.016 (3.506) (1.892) (0.048) (-0.346) (3.509) (1.351)
Diversification -0.012** -0.008 0.003 0.005 -0.014** -0.009 (-2.151) (-1.225) (0.311) (0.431) (-2.144) (-1.119)
Hostile 0.003 0.017 -0.003 -0.008 0.008 0.038 (0.177) (1.010) (-0.148) (-0.368) (0.421) (1.436)
Tender 0.002 0.016* -0.001 0.015 0.003 0.016 (0.345) (1.864) (-0.047) (0.935) (0.465) (1.542)
Relative Size -0.012** -0.019*** 0.017 0.019 -0.014** -0.022*** (-2.142) (-2.667) (0.907) (0.905) (-2.230) (-2.666)
Bidder ln(MV) -0.005*** -0.004*** -0.008** -0.010** -0.004** -0.003 (-3.848) (-2.629) (-2.324) (-2.540) (-2.508) (-1.577)
Bidder MTBV 0.001 0.001 0.000 0.002 0.001 0.001 (0.962) (1.297) (0.170) (1.013) (0.938) (0.820)
Bidder RunUps -0.015** -0.008 -0.017 -0.012 -0.016** -0.010 (-2.173) (-0.971) (-1.341) (-0.827) (-2.007) (-1.043)
Constant 0.029** 0.074*** 0.046 0.106* 0.025 0.082*** (2.128) (3.465) (1.199) (1.903) (1.635) (2.691)
N 606 606 155 155 451 451
R2 0.079 . 0.074 . 0.105 .
Hausman test X2 = 12.57
(p = 0.0004)
First-stage F test X2 = 13.23
(p = 0.0000)
, 1 , ,52i t i t i tOfferPremiums weekhigh
Chapter 3. Tables & Figures
145
Table 3.8: Robustness test: Controlling for other relevant price measures
This table presents the piecewise regression results for offer premiums on the target 52-week high, controlling for
other price measures that might be used as the reference point prices. Column (1) reports the relation between
offer premiums and the target 52-week high. Column (2) reports the results by controlling for the target 52-week
minimum (52wMin), which is the target lowest stock prices over the recent 52 weeks, which is measured as
logarithmic term difference between target lowest stock price over 335 calendar days ending 30 days prior to the
announcement date and the target stock price 30 days prior to the announcement date, column (3) reports the
results by controlling for the target 52-week mean (52wMean), which with a similar measure to the target 52-
week high and 52-week minimum. Robustness t-statistics are reported in parentheses. Statistical significance at
the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
Offer Premiums Other Price Measures
(1) (2) (3)
Low52wh 0.322*** 0.313*** 0.265**
(3.618) (3.224) (2.394)
Mid52wh 0.122 0.122 0.092
(1.057) (1.053) (0.766)
High52wh 0.150 0.150 0.109
(0.451) (0.448) (0.326)
52wMin 0.010
(0.310)
52wMean 0.055
(0.890)
Constant 0.229*** 0.234*** 0.243***
(16.271) (10.507) (11.747)
N 606 606 606
R2 0.055 0.056 0.057
Chapter 3. Tables & Figures
146
Table 3.9: Robustness test: Controlling for target past returns
This table presents the piecewise regression results for offer premiums on the target 52-week high, controlling for
other price measures that used as the reference points. Column (1) reports the regression results by controlling for
segments of the target 52-week high, which is the target highest stock price over 335 calendar days ending 30
days prior to the announcement date and the target stock price 30 days prior to the announcement date, column
(2) reports the results by controlling for the target past returns in one, two and three months prior to the takeover
announcement date, column (3) controls for the target past returns up to 6 months prior to the takeover
announcement date, column (4) controls for the target past returns up to one year prior to the takeover
announcement date. Robustness t-statistics are reported in parentheses. Statistical significance at the 1% level, 5%
level, and 10% level, denoted ***, ** and * respectively.
Offer Premiums (1) (2) (3) (4)
Low52wk 0.322*** 0.325*** 0.336*** 0.298**
(3.618) (3.413) (3.376) (2.388)
Mid52wk 0.122 0.104 0.115 0.114
(1.057) (0.868) (0.941) (0.742)
High52wk 0.150 0.150 0.164 -0.206
(0.451) (0.452) (0.428) (-0.392)
Target return t-1 0.038 0.034 0.005
(0.540) (0.467) (0.065)
Target return t-2 -0.078 -0.065 -0.108
(-1.012) (-0.803) (-1.101)
Target return t-3 0.036 0.016 0.027
(0.509) (0.217) (0.286)
Target return t-4 0.103 0.094
(1.288) (0.884)
Target return t-5 -0.070 -0.055
(-0.813) (-0.486)
Target return t-6 -0.111 -0.082
(-1.341) (-0.710)
Target return t-7 -0.103
(-1.040)
Target return t-8 -0.140
(-1.422)
Target return t-9 -0.112
(-1.314)
Target return t-10 0.091
(1.050)
Target return t-11 -0.132
(-1.309)
Target return t-12 -0.121
(-1.355)
Constant 0.229*** 0.229*** 0.229*** 0.236***
(16.271) (14.332) (13.818) (10.927)
N 606 602 573 418
R2 0.055 0.059 0.066 0.070
Chapter 3. Tables & Figures
147
Table 3.10: Robustness test: Reference point effect on offer premiums by market-wide valuation
This table presents the piecewise regression results for offer premiums on the target 52-week high by market
valuation. High Market is a dummy variable, taking a value of 1 if takeover months in the top 25% above past 5-
year average de-trended P/E of the UK market or market valuation is high, 0 otherwise. Specifications (1)-(3)
report the results when market-wide valuation is high (i.e. High), low (i.e. Low) and neutral (i.e. Neutral)
respectively. Details of this methodology were presented in methodology section of the Chapter 4. Other variable
definitions are as in the notes of Table 3.3. Equation is presented in the note of Table 3.5. Robustness t-statistics
are reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and
* respectively, is reported alongside the coefficients.
(1) (2) (3)
High Low Neutral
Offer premiums Coef. t-stat. Coef. t-stat. Coef. t-stat.
Low52wh 0.494*** (4.322) 0.761*** (2.704) 0.119 (0.725)
Mid52wh 0.181 (1.156) 0.136 (0.517) 0.431* (1.709)
High52wh 0.297 (0.683) -0.245 (-0.365) -0.669 (-0.818)
Stock -0.042 (-1.417) -0.066 (-1.063) -0.164*** (-3.354)
Cash -0.021 (-0.784) -0.079 (-1.492) -0.162*** (-3.843)
Diversification 0.005 (0.235) 0.035 (0.951) -0.053* (-1.772)
Hostile 0.065 (1.339) -0.032 (-0.308) 0.112* (1.807)
Tender 0.029 (1.136) 0.045 (1.041) 0.017 (0.465)
Cross Border 0.049* (1.954) 0.068 (1.410) 0.000 (0.004)
Relative Size 0.080*** (3.260) 0.058 (1.013) 0.030 (1.199)
Target ln(MV) -0.073*** (-7.514) -0.038** (-2.063) -0.070*** (-4.785)
Target MTBV -0.006** (-2.589) 0.004 (0.856) -0.006 (-1.183)
Target RunUps 0.098** (2.397) 0.071 (0.864) 0.071 (1.189)
Target Volatility -2.925*** (-3.026) -3.282*** (-2.988) -2.505 (-1.407)
Bidder ln(MV) 0.055*** (5.833) 0.030** (2.197) 0.060*** (4.093)
Bidder MTBV 0.000 (0.037) 0.001 (0.207) 0.006* (1.882)
Bidder RunUps 0.042 (1.587) -0.041 (-1.172) 0.041 (0.883)
Constant 0.227*** (4.113) 0.184* (1.820) 0.383*** (3.326)
N 300 110 196
R2 0.324 0.291 0.319
Chapter 3. Tables & Figures
148
Table 3.11: Robustness test: U.K. and U.S. bidder subsamples
This table reports the piecewise linear regression results for offer premiums on 52WeekHigh of both U.K. and U.S.
bidders’ acquisitions of U.K. targets. We document three individual target 52WeekHigh levels: Low52wh (0-30%),
Mid52wh (30%-70%), and High52wh (70% and above). Column (1) reports the results for U.K. domestic
acquisitions. Column (2) reports the results for U.S. bidders’ acquisitions of U.K. targets. Variable definitions are
as in the notes of Table 3.3. Equation is presented in the note of Table 3.5. Robustness t-statistics are reported in
parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively,
is reported alongside the coefficients.
(1) (2)
Offer premiums U.K. Domestic Cross-Border Subsample (U.S. bidders)
Coef. t-stat. Coef. t-stat.
Low52wk 0.404*** (4.191) 0.593** (2.488)
Mid52wk 0.093 (0.768) 0.376 (0.943)
High52wk 0.186 (0.469) -0.661 (-0.633)
Stock -0.072*** (-2.963) -0.296** (-2.547)
Cash -0.077*** (-3.285) -0.096 (-0.861)
Diversification -0.002 (-0.097) -0.032 (-0.537)
Hostile 0.157*** (2.748) -0.149 (-1.446)
Tender 0.008 (0.377) 0.141*** (2.773)
Relative Size 0.047** (2.451) 0.242 (1.158)
Target ln(MV) -0.069*** (-8.700) -0.053* (-1.976)
Target MTBV -0.004** (-2.232) -0.009 (-0.986)
Target RunUps 0.050 (1.370) 0.166 (1.585)
Target Volatility -2.894*** (-3.576) -6.084*** (-2.822)
Bidder ln(MV) 0.054*** (7.266) 0.047 (1.442)
Bidder MTBV 0.001 (0.666) 0.025** (2.141)
Bidder RunUps 0.024 (1.014) 0.003 (0.046)
Constant 0.291*** (5.562) 0.239 (1.045)
N 451 67
R2 0.274 0.453
Chapter 3. Tables & Figures
149
Table 3.12: Robustness test: Reference point effect on offer premiums by the method of payment
This table presents the piecewise linear regression results of offer premiums on the target 52-week high in different
payment method subsamples. Column (1) includes acquisitions that are purely financed by Cash, column (2)
includes those that are purely financed by stocks and column (3) includes those financed by a mixture of cash and
stocks. Other control variable definitions are as in the notes of Table 3.3. Equation is presented in the note of Table
3.5. Robustness t-statistics are reported in parentheses. Statistical significance at the 1% level, 5% level, and 10%
level, denoted ***, ** and * respectively, is reported alongside the coefficients.
(1) (2) (3)
Offer premiums Cash Stocks Mixed
Coef. t-stat. Coef. t-stat. Coef. t-stat.
Low52wh 0.420*** (3.557) 0.397** (2.138) 0.378* (1.774)
Mid52wh 0.248 (1.490) 0.150 (0.688) 0.348* (1.666)
High52wh 0.319 (0.661) 0.292 (0.432) -0.891 (-1.557)
Diversification 0.016 (0.786) -0.060* (-1.784) 0.015 (0.459)
Hostile -0.006 (-0.094) 0.278*** (3.813) 0.169*** (2.801)
Tender 0.055** (2.496) -0.053 (-1.188) -0.063 (-1.021)
Cross Border 0.045* (1.912) 0.016 (0.245) 0.022 (0.433)
Relative Size 0.056* (1.829) 0.061 (1.505) 0.102** (2.324)
Target ln(MV) -0.049*** (-5.431) -0.087*** (-3.812) -0.114*** (-4.953)
Target MTBV -0.005** (-2.428) 0.001 (0.106) -0.004 (-1.100)
Target RunUps 0.092** (2.329) 0.063 (0.889) 0.151** (2.042)
Target Volatility -3.632*** (-4.064) -1.707 (-1.318) -1.280 (-0.605)
Bidder ln(MV) 0.041*** (5.060) 0.052** (2.305) 0.100*** (4.415)
Bidder MTBV 0.004 (1.544) 0.000 (0.011) 0.002 (0.750)
Bidder RunUps 0.018 (0.582) 0.055 (1.486) -0.082 (-1.480)
Constant 0.172*** (3.332) 0.334*** (4.129) 0.222* (1.835)
N 380 108 118
R2 0.269 0.350 0.341
Chapter 3. Tables & Figures
150
Table 3.13: Robustness test: Reference point effect on offer premiums for diversified and undiversified
M&A deals
This table presents the piecewise linear regression results of offer premiums on the target 52-week high in different
payment method subsamples. Column (1) includes only diversified acquisitions (i.e. Diversificiation), column (2)
includes only undiversified acquisitions (i.e. Relatedness). Other control variable definitions are as in the notes of
Table 3.3. Equation is presented in the note of Table 3.5. Robustness t-statistics are reported in parentheses.
Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively, is reported
alongside the coefficients.
(1) (2)
Offer premiums Diversification Relatedness
Coef. t-stat. Coef. t-stat.
Low52wh 0.468*** (3.515) 0.377*** (3.155)
Mid52wh 0.187 (1.261) 0.261 (1.551)
High52wh 0.168 (0.393) -0.115 (-0.213)
Stock -0.097*** (-3.032) -0.064* (-1.891)
Cash -0.061** (-2.003) -0.079** (-2.587)
Hostile 0.041 (1.028) 0.123 (1.592)
Tender 0.019 (0.727) 0.034 (1.309)
Cross Border 0.023 (0.843) 0.063** (2.155)
Relative Size 0.027 (1.039) 0.053* (1.961)
Target ln(MV) -0.046*** (-3.970) -0.071*** (-6.489)
Target MTBV -0.002 (-0.759) -0.007*** (-2.734)
Target RunUps 0.072 (1.581) 0.107** (2.393)
Target Volatility -2.444*** (-2.595) -3.757*** (-3.538)
Bidder ln(MV) 0.037*** (3.662) 0.051*** (4.731)
Bidder MTBV 0.002 (0.837) 0.002 (0.700)
Bidder RunUps 0.008 (0.237) 0.035 (1.280)
Constant 0.264*** (4.323) 0.321*** (4.576)
N 292 314
R2 0.200 0.323
Chapter 3. Tables & Figures
151
Table 3.14: Robustness test: Reference point effect on offer premiums for tender offers and mergers
This table presents the piecewise linear regression results of offer premiums on the target 52-week high in different
payment method subsamples. Column (1) includes only tender offers (i.e. Tender), column (2) includes only
mergers (i.e.Merger). Other control variable definitions are as in the notes of Table 3.3. Equation is presented in
the note of Table 3.5. Robustness t-statistics are reported in parentheses. Statistical significance at the 1% level,
5% level, and 10% level, denoted ***, ** and * respectively, is reported alongside the coefficients.
(1) (2)
Offer premiums Tender Merger
Coef. t-stat. Coef. t-stat.
Low52wh 0.410*** (4.167) 0.402** (2.389)
Mid52wh 0.218* (1.718) 0.151 (0.670)
High52wh -0.295 (-0.809) 0.846 (1.143)
Stock -0.086*** (-3.557) -0.122* (-1.693)
Cash -0.070*** (-3.307) -0.103 (-1.497)
Diversification -0.021 (-1.162) 0.035 (1.108)
Hostile 0.078* (1.793) 0.113*** (2.925)
Cross Border 0.059** (2.482) -0.000 (-0.005)
Relative Size 0.047** (2.296) 0.186** (2.393)
Target ln(MV) -0.070*** (-7.117) -0.053*** (-4.546)
Target MTBV -0.003 (-1.209) -0.006** (-2.479)
Target RunUps 0.051 (1.374) 0.147*** (2.710)
Target Volatility -3.034*** (-3.916) -2.193 (-1.478)
Bidder ln(MV) 0.051*** (5.689) 0.043*** (3.950)
Bidder MTBV 0.001 (0.763) 0.003 (0.683)
Bidder RunUps 0.025 (1.023) -0.007 (-0.179)
Constant 0.330*** (7.075) 0.250** (2.312)
N 408 198
R2 0.280 0.247
Chapter 3. Tables & Figures
152
Table 3.15: Robustness test: Market model for 5-day CARs
This table presents both OLS and 2SLS regression results of bidder market model 5-day announcement returns
(CAR5mm) on the offer premium. We estimate market model parameters over the window from 261 to 28 trading
days prior to the announcement date [-261,-28]. The target 52-week high is used as an instrument. We report the
regressions for both cross-border and domestic acquisitions into the United Kingdom. Hausman test and first-
stage F-test are reported below the table. Variable definitions are as in the notes of Table 3.3. The 2SLS equation
is presented in the note of Table 3.7. Robustness t-statistics are reported in parentheses. Statistical significance at
the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively. (1) (2) (3) (4) (5) (6)
CAR5mm Full Sample Cross-Border Bidders U.K. Bidders
OLS IV OLS IV OLS IV
Offer Premiums -0.016 -0.183*** 0.022 -0.152 -0.036** -0.210** (-1.155) (-2.982) (0.866) (-1.551) (-2.135) (-2.302)
Stock 0.005 -0.004 -0.003 -0.021 0.006 -0.004 (0.575) (-0.393) (-0.139) (-0.689) (0.597) (-0.352)
Cash 0.027*** 0.019** 0.008 -0.002 0.030*** 0.019* (3.573) (2.103) (0.399) (-0.089) (3.544) (1.707)
Diversification -0.011* -0.007 -0.002 0.000 -0.011* -0.007 (-1.950) (-1.145) (-0.162) (0.029) (-1.730) (-0.958)
Hostile 0.003 0.016 -0.003 -0.009 0.010 0.033 (0.207) (0.967) (-0.147) (-0.389) (0.500) (1.337)
Tender -0.001 0.011 -0.005 0.013 0.000 0.010 (-0.171) (1.343) (-0.421) (0.769) (0.036) (1.041)
Relative Size -0.012** -0.018*** 0.016 0.019 -0.014** -0.020*** (-2.118) (-2.618) (0.844) (0.838) (-2.283) (-2.657)
Bidder ln(MV) -0.006*** -0.005*** -0.009*** -0.012*** -0.004*** -0.004* (-4.267) (-3.130) (-2.625) (-2.793) (-2.708) (-1.945)
Bidder MTBV 0.001 0.001 0.001 0.004 0.001 0.001 (1.164) (1.457) (0.550) (1.383) (1.048) (0.958)
Bidder RunUps -0.040*** -0.033*** -0.036*** -0.030* -0.043*** -0.039*** (-5.934) (-4.321) (-2.709) (-1.924) (-5.527) (-4.274)
Constant 0.032** 0.072*** 0.050 0.120** 0.028* 0.072** (2.407) (3.501) (1.258) (2.030) (1.850) (2.554)
N 606 606 155 155 451 451
R2 0.134 . 0.123 . 0.161 .
Hausman test X2 = 10.02
(p=0.0016)
First-stage F test X2= 13.23
(p= 0.0000)
Chapter 3. Tables & Figures
153
Table 3.16: Robustness test: Market-adjusted model for 3-day CARs
This table presents both OLS and 2SLS regression results of bidder market-adjusted model 3-day announcement
returns (CAR3ma) on the offer premium. The target 52-week high is used as an instrument. We report the
regressions for both cross-border and domestic acquisitions into the United Kingdom. Hausman test and first-
stage F-test are reported below the table. Variable definitions are as in the notes of Table 3.3. The 2SLS equation
is presented in the note of Table 3.7. Robustness t-statistics are reported in parentheses. Statistical significance at
the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
(1) (2) (3) (4) (5) (6)
CAR3ma Full Sample Cross Border Bidders U.K. Bidders
OLS IV OLS IV OLS IV
Offer Premiums -0.017 -0.175*** 0.017 -0.105 -0.031** -0.220**
(-1.304) (-3.191) (0.899) (-1.332) (-2.131) (-2.578)
Stock 0.009 0.000 0.006 -0.006 0.009 -0.002
(0.993) (0.029) (0.211) (-0.248) (0.966) (-0.139)
Cash 0.028*** 0.020** 0.010 0.003 0.031*** 0.019*
(3.809) (2.557) (0.469) (0.171) (3.912) (1.811)
Diversification -0.010** -0.007 0.006 0.007 -0.014** -0.009
(-2.096) (-1.215) (0.666) (0.699) (-2.357) (-1.343)
Hostile 0.002 0.014 0.001 -0.003 0.006 0.031
(0.185) (0.968) (0.077) (-0.169) (0.396) (1.355)
Tender 0.004 0.015** -0.003 0.010 0.007 0.017*
(0.721) (2.110) (-0.233) (0.743) (1.069) (1.926)
Relative Size -0.008 -0.013** 0.012 0.013 -0.009 -0.015**
(-1.195) (-2.133) (0.493) (0.731) (-1.383) (-2.180)
Bidder ln(MV) -0.003** -0.002 -0.004* -0.006* -0.002 -0.001
(-2.424) (-1.526) (-1.720) (-1.841) (-1.338) (-0.693)
Bidder MTBV 0.000 0.001 0.000 0.002 0.000 0.000
(0.579) (1.002) (0.078) (0.927) (0.566) (0.532)
Bidder RunUps -0.013 -0.007 -0.016 -0.012 -0.013 -0.008
(-1.582) (-1.025) (-1.100) (-0.973) (-1.389) (-0.951)
Constant 0.009 0.047** 0.014 0.063 0.006 0.055**
(0.780) (2.541) (0.384) (1.316) (0.505) (2.063)
N 606 606 155 155 451 451
R2 0.075 . 0.055 . 0.102 .
Hausman test X2 = 7.02
(p=0.0083)
First-stage F test X2= 6.41
(p= 0.0003)
Chapter 3. Tables & Figures
154
Table 3.17: Robustness test: Market model for 3-day CARs
This table presents both OLS and 2SLS regression results of bidder market model 3-day announcement returns
(CAR3mm) on the offer premium. We estimate market model parameters over the window from 261 to 28 trading
days prior to the announcement date [-261,-28]. The target 52-week high is used as an instrument. We report the
regressions for both cross-border and domestic acquisitions into the United Kingdom. Hausman test and first-
stage F test are reported below the table. Variable definitions are as in the notes of Table 3.3. The 2SLS equation
is presented in the note of Table 3.7. Robustness t-statistics are reported in parentheses. Statistical significance at
the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
(1) (2) (3) (4) (5) (6)
CAR3mm Full Sample Cross Border Bidders U.K. Bidders
OLS IV OLS IV OLS IV
Offer Premiums -0.015 -0.151*** 0.026 -0.119 -0.033** -0.173**
(-1.133) (-2.848) (1.482) (-1.437) (-2.207) (-2.184)
Stock 0.007 -0.001 0.002 -0.013 0.007 -0.001
(0.768) (-0.066) (0.065) (-0.494) (0.766) (-0.073)
Cash 0.027*** 0.021*** 0.012 0.004 0.030*** 0.021**
(3.829) (2.731) (0.629) (0.231) (3.930) (2.233)
Diversification -0.010** -0.007 0.001 0.003 -0.012** -0.009
(-2.078) (-1.315) (0.129) (0.270) (-2.094) (-1.353)
Hostile 0.002 0.013 0.004 -0.001 0.005 0.024
(0.220) (0.899) (0.264) (-0.071) (0.348) (1.112)
Tender 0.002 0.011 -0.005 0.010 0.004 0.012
(0.315) (1.636) (-0.431) (0.723) (0.652) (1.422)
Relative Size -0.009 -0.013** 0.008 0.010 -0.010 -0.015**
(-1.355) (-2.297) (0.357) (0.538) (-1.542) (-2.257)
Bidder ln(MV) -0.004*** -0.003** -0.006** -0.008** -0.002 -0.002
(-3.042) (-2.150) (-2.101) (-2.229) (-1.584) (-1.056)
Bidder MTBV 0.001 0.001 0.001 0.003 0.000 0.000
(0.902) (1.283) (0.452) (1.432) (0.762) (0.745)
Bidder RunUps -0.026*** -0.021*** -0.026* -0.021 -0.027*** -0.024***
(-3.154) (-3.138) (-1.741) (-1.646) (-2.921) (-3.012)
Constant 0.014 0.047*** 0.018 0.077 0.011 0.047*
(1.285) (2.628) (0.515) (1.536) (0.866) (1.889)
N 606 606 155 155 451 451
R2 0.108 . 0.093 . 0.135 .
Hausman test X2 = 3.96
(p=0.0471)
First-stage F test X2= 6.41
(p= 0.0003)
Chapter 3. Tables & Figures
155
Figure 3.1: Distribution of number of deals by bidders
This figure plots the time-series distribution of the number of deals for the full sample, the U.K. domestic sample,
and the CrossBorder sample. The U.K. domestic sample contains U.K. public bidders acquire U.K. public targets;
The Cross-Border sample contains those foreign public bidders acquire U.K. public targets. Y-axis indicates the
number of deals, X-axis represents year, spanning from 1985 to 2014.
0
10
20
30
40
50
60
70
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
201
2
201
3
201
4
Nu
mb
er o
f d
eals
All UK CrossBorder
Chapter 3. Tables & Figures
156
Figure 3.2: Distribution of value of deals by bidder types This figure plots the time-series distribution of the total value of deals for the full sample, the U.K. domestic
sample, and the CrossBorder sample. The U.K. domestic sample contains U.K. public bidders acquire U.K. public
targets; The CrossBorder sample contains those foreign public bidders acquire U.K. public targets. Y-axis indicates
the value of deals in billion U.S. dollars, X-axis represents year, spanning from 1985 to 2014.
0
10
20
30
40
50
60
701
98
5
198
7
198
9
199
1
199
3
199
5
199
7
199
9
200
1
200
3
200
5
200
7
200
9
201
1
201
3
To
tal
va
lue
of
dea
ls i
n $
bil
All UK CrossBorder
Chapter 3. Tables & Figures
157
Figure 3.3: Offer premiums relative to the target 52-week high
This figure presents the density of the difference between offer premiums and the target 52-week high, where
Offer Premiums is the logarithmic term difference between the offer price and the target stock price 30 days prior
to the takeover announcement and 52WeekHigh is the logarithmic term difference between the target’s highest
stock prices over the 335 calendars ending 30 days prior to the announcement date and the target stock price 30
days prior to the announcement date.
0.5
11.5
2
De
nsity
-5 5OfferPremiums - 52WeekHigh (%)
Chapter 3. Tables & Figures
158
Figure 3.4: Nonlinear relation between offer premiums and the target 52-week high
This figure presents the nonlinear effect of the target 52-week high on offer premiums, where Offer Premiums is
the logarithmic term difference between the offer price and the target stock price 30 days prior to the takeover
announcement and 52WeekHigh is the logarithmic term difference between the target highest stock price over the
335 calendars ending 30 days prior to the announcement date and the target stock price 30 days prior to the
announcement date. We use local polynomial regression to smooth scatter plots, and set our sample where both
offer premiums and the target 52-week high are larger than 0, and less than 100%.
25
30
35
40
45
Offe
r P
rem
ium
s (
%)
0 10 20 30 40 50 60 70 80 90 10052 Week High (%)
159
CHAPTER 4: RELATIVE
REFERENCE PRICES AND M&A
MISVALUATIONS
Chapter 4. Relative Reference Prices and M&A Misvaluations
160
Abstract
This chapter examines the misvaluation hypothesis using a relative reference point (RRP) in the M&A
market. The reference point is the deviation of a firm’s current stock price from its 52-week high,
which reflects investors’ perception of the firm’s valuation. The market perceives the firm to be
overvalued when its current price is close to the 52-week high and to be undervalued when the current
price falls below its 52-week high significantly. The RRP indicates the extent to which the bidder is
more overvalued relative to the target. The sign of bidders’ overvaluation is revealed especially if
bidders pay takeover targets with stocks and with higher offer premiums. It was found that the RRP
accommodates the implications with respect to the misvaluation hypothesis. The results of this
chapter show that bidders prefer stock payments when the RRP increases, indicating that bidders are
overvalued and use their stocks to accelerate the process of overvaluation dilution. It became clear
that the RRP is positively related to the offer premium, suggesting that a relatively more overvalued
bidder tends to overpay for the target for the purpose of diluting overvaluation. With regard to the
relationship between the RRP and M&A outcomes, it became apparent that the RRP is found to be
positively related to target announcement returns, and is negatively related to the bidder
announcement returns. It also became apparent that the RRP is positively related to the stock bidders’
long-term abnormal returns. This result shows evidence of bidders’ rationality, eliminating concerns
about paying according to the RRP results in underperformance. Thus, the results are consistent with
the predictions of the misvaluation hypothesis and reference point theory.
Chapter 4. Relative Reference Prices and M&A Misvaluations
161
4.1. Introduction
“Why could the typical investor expect any better success in trying to buy at low levels and sell at
high levels than in trying to forecast what the market is going to do? Because if he does the former
he acts only after the market has moved down into buying levels or up into selling levels. His role is
not that of a prophet but of a businessman seizing clearly evident investment opportunities. He is not
trying to be smarter than his fellow investors but simply trying to be less irrational than the mass of
speculators who insist on buying after the market advances and selling after it goes down. If the
market persists in behaving foolishly, all he seems to need is ordinary common sense in order to
exploit its foolishness.”
Benjamin Graham (1949: 31)
The misvaluation hypothesis explains an important motive of merger and acquisition (M&A)
activities. The theoretical model of Shleifer and Vishny (2003) predicts that the stock market drives
M&As. Overvalued bidders who serve the long-term interests of the shareholders will dilute
overvaluation through stock-financed acquisitions, as using stocks would accelerate such a dilution
process. The misvaluation hypothesis holds that bidders are rational whereas the market is irrational,46
which violates the efficiency market hypothesis and differs from what Roll’s (1986) hubris
management predicts. Following Shleifer and Vishny (2003), Dong et al. (2006) provided direct
evidence that bidders overpay for targets as long as bidders are overvalued relative to targets. Ang
and Cheng (2006), who investigated the long-term performance of stock bidders, found a positive
46 Shleifer and Vishny (2003) suggest that bidders are rational as they could time the market and use overvalued stocks
as a means of payment for acquisitions rather than holding them until they are corrected in the market. Bidders would pay
a price that is lower than the estimated synergy created by the combination of firms. The model assumes a target and a
bidder with stock volumes K and K1 respectively, price per unit Q and Q1 (where Q1>Q) and price per unit for the combined
equity S. As synergies can be achieved when S(K+K1)-(KQ+K1Q1)>0 the bidders would pay a price P to gain synergies
as long as it lies between Q and S, or Q<P<S.
Chapter 4. Relative Reference Prices and M&A Misvaluations
162
relationship between overvalued stocks and long-term performance. Rhodes-Kropf and Viswanathan
(2004) provided a behavioural model suggesting that fully rational individuals make mistakes as they
tend to overestimate synergies, especially when market-wide valuation is high (Rhodes-Kropf et al.,
2005).
Conventional misvaluation measures face three major challenges. First, the measures relating to a
firm’s fundamental value cause estimation biases. Different firms could measure their assets in
different accounting approaches, such as adopting fair value and historical cost approaches vary
across the firms, which lead to variation of the firm’s fundamental value. Companies are also found
to manipulate accounting figures to raise the firm’s value, which is especially prevalent prior to the
financial crisis period. Secondly, existent misvaluation measures are mainly based on historical or
forward-looking information, such as price-to-book value (P/B), price-to-residual income value (P/V)
and earnings per share (EPS), which are less likely to reflect the latest status of the firm. Thirdly, the
frequently used ratio of MTBV is a proxy for both mispricing and investment opportunities of the
firm. According to Di Giuli (2013)47, firms with better investment opportunities should also increase
the practice of using stocks in acquisitions, leading to the same prediction as the misvaluation
hypothesis.
This chapter constructs a novel misvaluation measure: the relative reference point (RRP), which is
derived from the reference point effect48. Baker et al. (2012) defined the target reference point as the
47 Di Giuli (2013) proposed some post-merger investment-related proxies to disentangle the effects of mispricing and
investment opportunities.
48 Kahneman and Tversky (1979) proposed that with the reference point effect that people rely heavily on a single salient
piece of information while making decisions. Prior research has highlighted the 52-week high as a reference price ( Kliger
and Kudryavtsev, 2008; Barberis and Xiong, 2009).
Chapter 4. Relative Reference Prices and M&A Misvaluations
163
deviation of a target’s current stock price from its 52-week high.49 Following this, the bidder reference
point in this chapter is defined as the deviation of a bidder’s current stock price from its 52-week high,
reflecting the extent to which the bidder is overvalued. It is argued that the firm’s 52-week high offers
the market an insight that the firm is overvalued, given that a firm’s 52-week high is the outcome of
a series of good news occurred in the past, the price will move toward to the firm’s fundamental value
once the price momentum disappears. When a firm’s current price is close to its highest stock price
over the year prior to takeover announcement, it suggests that the firm is still strong in price
momentum and should be reluctant to push up the price on the rationale that the firm is already
overvalued, whilst if a firm’s current price falls significantly below the 52-week high, substantiating
the market’s belief that the 52-week high is a sign of overvaluation. Based on this, the RRP is defined
as the difference between the target and the bidder reference points, indicating the extent to which the
bidder is relatively more overvalued than the target (i.e. the relatively more overvalued bidders).50
Baker et al. (2012) realised that the offer premium is positively related to the target reference point,
as the target uses its reference point to reinforce the bargaining position and demand high offer
premiums in an M&A deal, especially when the current price deviates significantly from its 52-week
high.51 On the other hand, bidders paying offer premiums based on the target 52-week high would
argue that they can outperform the target’s recent high and could generate synergies for the combined
firm following an M&A deal. In order to achieve the deal, bidders tend to pay for the target according
49 Their paper used’ the target reference point’ and ‘the target 52-week high’ interchangeably. In this chapter, the reference
point refers to the difference between the firm’s highest stock price over the period of 52 weeks prior to the M&A
announcement (52-week high) and the firm’s current price at the M&A announcement date. Such an approach makes this
reference point easier to interpret.
50 The construction of RRP is presented in the methodology section of this chapter.
51 They found a nonlinear relationship between the offer premium and the target 52-week high. In each segment, the offer
premium increases with the target 52-week high.
Chapter 4. Relative Reference Prices and M&A Misvaluations
164
to its 52-week high. Their data support this view, finding that an offer price paid according to the
target 52-week high increase the deal success rate.
However, the idea of investigating the target 52-week high effect on the offer premium can explain
the negotiation process between the management teams of the two firms and how the firm is priced.
Baker et al. (2012) failed to reveal the bidders’ M&A motive as one cannot tell whether the proposed
offer price is a result of strangeness of the bargaining power of the target or any other considerations
of bidders. Their paper shows that bidders who believe that paying for acquisitions according to the
target 52-week high can generate wealth for shareholders. As a matter of fact, it destroys their wealth,
reflected in negative bidders’ announcement returns. Negative market reactions have been found
when bidders paying for acquisitions according to the target reference point results from overpayment.
Unlike those market investors, who are less experienced or lacking of information about the firms,
bidder managers can have more information about target’s value, and thus their M&A pricing
decisions should not be based on a single price number. It leads us to the question of whether bidders
have a similar view with the market with respect to the reference point. Specifically, the market might
believe that the target 52-week high represents a sign of poor performance when the target’s current
price falls below the 52-week high significantly, since it is less likely to rebound to the 52-week high,
whereas bidders paying for takeover targets according to the target reference point might believe that
the target is currently undervalued. It is evident that assessing a firm’s short-term performance does
not enable direct observation. Rather than short-selling the target’s stocks, bidders paying offer
premiums according to the target reference point have confidence in managing the deal afterwards.
Therefore, it can be argued that offer premiums based on the reference point are to guarantee
acceptance of the deal.
Chapter 4. Relative Reference Prices and M&A Misvaluations
165
Another important question is whether the market also takes a look at the bidder reference point for
a similar reason to that for the target reference point, as this is also readily available to the public.
Simple intuition suggests that the market believes the bidder 52-week high is a highly relevant piece
of price information about the target value and treats it as a reference point, just like the target 52-
week high. The bidder reference point offers bidder managers an insight into how much they are able
to pay for the target. The bidder believes that the target 52-week high reflects the firm’s potential
profit-generating ability, and pays M&A offer premiums based on this, with the incentive of restoring
targets’ price discounts. On the other hand, target managers regard the bidder reference point as how
much in offer premiums they can possibly negotiate for when selling the firm. Baker et al. (2012)
found that targets demand high offer premiums based on the target 52-week high, relying on the
assumption that targets are not overvalued, otherwise they may be less likely to sell the firm for a
higher offer premium if they are undervalued. Therefore, the RRP reflects the market perception of a
firms’ valuation. The proxy explains relative valuations from the market angle, eliminating concerns
about any biases of a firm’s fundamental value.
Since both the target and the bidder reference point have a value in explaining how M&As are
structured, solely investigating the target reference point or the bidder reference point does not fully
account for the reference point effect on the M&A valuation. It should be expected that the market
will look at the bidder reference point believing that bidders whose current price is close to the 52-
week high are likely to create value given that they have performed well recently. This could also
reinforce the bargaining power of the bidders in an M&A deal, resulting in a lower offer premium
paid for the target, similar to Chira and Madura’s view (2015). In this situation, when bidders pay for
targets with a high offer premium, this would lead the market to believe that there is an overpayment.
Chapter 4. Relative Reference Prices and M&A Misvaluations
166
However, according to the misvaluation hypothesis, managers could time the market by paying with
overvalued stocks for acquisitions. If the RRP is a suitable representation for relative valuations,
bidders looking at reference points of the two firms involved are actually going to time the market. If
this is the case, it remains largely unexplored why bidders rely on their reference point while paying
for the target, given better understanding about the firm.
The RRP is able to capture the market perception of the misvaluation of the two firms involved in an
M&A deal. Unlike Dong et al. (2006) and Rhodes-Kropf and Viswanathan (2004) who provide a
different view of bidder and target’s misvaluation,52 this researcher provides direct evidence that the
reference point should unify the investors’ view on the misvaluation of the two firms involved. George
and Hwang (2004) find misvaluation is driven by investors’ reactions. Lacking private information,
investors are reluctant to bid up a stock price when it is close to the 52-week high, based on the
rationale that it is previous good news that has driven a firm’s value beyond its fundamental value,
leading them to believe that a firm maybe overvalued. In contrast, investors should also be reluctant
to sell stocks of a firm whose price is far below the 52-week high, as this indicates that the firm maybe
undervalued. Extending their argument in the context of M&As, it becomes possible that managers
are theoretically highly committed in creating long-term value for the firm instead of enjoying short-
term profits from possible mispricing phenomena. Rather than Dong et al. (2006) who used the
difference between bidder’s and target’s P/B and P/V as proxies for the relative valuations between
the two firms, which are based on the firms’ fundamental value and argued that there is no actual
misvaluation but rather a managerial perception of misvaluation, the case of possible misvaluation
52 Dong et al. (2006) proposed price-to-book value and price-to-residual income value for bidder misvaluation whereas
Rhodes-Kropf and Viswanathan (2004) proposed a misvaluation measure based on an assumption that targets are rational,
and they overestimate merger synergies when valuation errors are sizable.
Chapter 4. Relative Reference Prices and M&A Misvaluations
167
with the RRP from a market perspective is examined.
To the best of this researcher’s knowledge, this is the first paper to examine the role of the bidder
reference point in the context of M&A valuation. Following Baker et al. (2012), who proposed the
target reference point to explain how much bidders should pay for the acquisition target, it is
suggested that the bidder reference point would lead bidders to consider how much they are able to
pay for the acquisition target. Bidders who have performed poorly recently may find it difficult to
provide any rationale to pay the target with high offer premiums, while bidders whose stock price is
close to its 52-week high should be regarded as rich in financing resources, and thus dominate the
negotiation table, resulting in a low offer premium. It is much easier for them to persuade their
shareholders that the firm is stable and they can manage the deal well afterwards. In addition, in order
to boost a firm’s value and retain the market’s confidence in the firm, bidders should initiate an M&A
bid by sticking to the reference point, a way the market can judge, while remaining silent when the
firm’s price is close to the 52-week high tends to be very risky for the firm’s prospects. Barberis and
Xiong (2009) suggest that investors tend to sell stocks whose value has recently risen. Grinblatt and
Kelaharju (2001) and Huddart et al. (2009) report large abnormal sales’ volumes around the 52-week
high.
Another reason that bidders look at their own reference point is simply because bidder reference point
is an important component for managers to time the market. It is expected that managers can exploit
mispricing, as in the prior M&A literature that bidders are likely to pay with stocks for acquisitions
when they are overvalued (Ang and Cheng, 2006; Dong et al., 2006). However, the proxies used in
their papers for misvaluation do not accurately reflect the information such as a firm’s prospects and
Chapter 4. Relative Reference Prices and M&A Misvaluations
168
managerial effectiveness, whilst the stock price is a relatively more comprehensive information.53
Therefore, it is suggested that the RRP, indicating the extent to which the market’s misperception of
the firm’s valuation, should provide similar findings as those predicted in the misvaluation hypothesis.
The RRP also signals to the target firm what price it could potentially negotiate with the bidder. If
targets believe, using the reference point, that they are overvalued, they would find it even harder to
justify this (overvaluation) given that they tend to be generally smaller and lack better investment
opportunities than bidders. This leads them to accept even more overvalued stocks for liquidity
purposes. In addition, an increase in the RRP also leads the target shareholders to believe that takeover
bidders are more attractive as it is more likely for the bidder to hit its reference price again more than
the target does. Hence, they might perceive that selling their firm to a well-run bidder would be more
likely to create value. According to Burch et al. (2012), targets tend to reserve bidders’ overvalued
stocks, maintaining that highly-valued bidders perform well or have better investment opportunities.
Therefore, stocks of the relatively more overvalued bidders are more attractive to targets.
Using M&As to investigate the relationship between the RRP and M&As is of great interest mainly
for two reasons. First, the RRP is a direct misvaluation measure that captures the market’s perception
of the firm’s valuation, avoiding biases caused by a firm’s fundamental characteristics (Lin et al.,
2011). M&As serving as a major corporate investment activity draw a great deal of investors’
attention. With limited information and limited time in which to process that information, investors
are likely to make decisions based on the most current market perceptions of a firm’s valuation,
53 The existence of mispricing in the short term has been confirmed even if the market is efficient, such as investors’
underreaction (Debondt and Thaler, 1985).
Chapter 4. Relative Reference Prices and M&A Misvaluations
169
making the RRP a suitable valuation proxy for this testing ground. Secondly, the RRP will facilitate
the M&A process. Bidders can identify the relative overvaluation through the RRP, as an increase of
the RRP would potentially drive a relatively more overvalued bidder to dilute overvaluation through
acquisitions.
Analysing a sample of 1,878 U.S. domestic public acquisitions announced between 1985 and 2014
and using the RRP to test the predictions of the misvaluation hypothesis, it was found that the
propensity to use stocks as a means of payment for acquisitions increases with the RRP, which is
more pronounced when market-wide valuation is high. The results of this chapter are robust after
applying additional controls. Moreover, the relatively more overvalued bidders tend to pay higher
offer premiums. The results continue to hold after endogeneity checks that may become necessary
due to omitted variable biases. Finally, multivariate analysis results show that the RRP plays a role in
bidders’ overpayment in the short term. However, it became evident that higher offer premiums paid
by the relatively more overvalued bidders are translated into less negative abnormal returns in the
long term, suggesting that bidders who pay for acquisitions according to the RRP protect the wealth
of shareholders. Overall, the findings in this chapter are consistent with the predictions of the
misvaluation hypothesis and reference point theory.
This chapter makes three distinct contributions to the literature. Firstly, it explains the misvaluation
hypothesis from the perspective of the reference point. A dynamic valuation framework with the RRP
is developed, based on market perception of a firm’s valuation. The RRP overcomes the effect of
market-wide valuation for an individual firm, as the two firms involved experience the same market
conditions. A firm’s current price that is close to its 52-week highest point is more likely to be driven
Chapter 4. Relative Reference Prices and M&A Misvaluations
170
by market-wide conditions (sentiment) and therefore associated with larger market valuation errors,
increasing the probability of the firm being overvalued. Using the RRP, it is possible to examine how
the valuation difference of the deviation of the two firms drives M&As. Hence, the bar is raised to
the market level, eliminating estimation biases arising from the use of a firm’s fundamental value.
Furthermore, the RRP is easily observable, which encourages investors to use it as a valuation
benchmark. The Results of This chapter suggest that the RRP is able to accommodate the implications
of the misvaluation hypothesis.
Secondly, the chapter provides direct evidence that more experienced investors behave similarly to
less experienced investors in major corporate investment decisions. However, the market and
managers may interpret the reference point effect differently as the market looks at the offer premium
paid according to the RRP as a result of overpayment by hubris-infected bidders, while bidders paying
for acquisitions according to the RRP is as a result of timing the market. The results indicate the less
negative long-term abnormal returns for the relatively more overvalued stock bidders compared with
the relatively less overvalued (or more undervalued) stock bidders.
Thirdly, this chapter offers a new insight into the method of payment hypothesis. It is suggested that
the sign of relative overvaluation is well identified by the RRP, which relaxes the assumptions of
irrational targets (Shleifer and Vishny, 2003) and valuation error misled targets (Rhodes-Kropf and
Viswanathan, 2004) related to the target’s motive of accepting overvalued stocks,54 as both bidders
and targets can identify any relative overvaluation in an M&A deal. Bidders paying stocks instead of
54 Common assumptions on whether the target will accept the overvalued stocks suggest that targets either have a cash-
out purpose (Shleifer and Vishny, 2003) or are misled by the market perception (Rhodes-Kropf and Viswanathan, 2004).
Chapter 4. Relative Reference Prices and M&A Misvaluations
171
cash for a larger RRP acquisition reduce offer premiums while paying cash in a lower RRP acquisition
reduce offer premiums compared with the case of the higher RRP acquisitions. Hence, the RRP
justifies the method of payment choice.
The remainder of this chapter is organised as follows: Section 2 provides a range of literature that is
related to the misvaluation hypothesis and reference point theory of M&As. Section 3 designs
hypotheses with regard to the relations between RRP and the method of payment, offer premiums and
M&A outcomes in the short and long runs. Section 4 summarises the data and presents the
methodologies. Section 5 analyses the empirical results. Section 6 conducts further robustness checks
regarding the role of the RRP played in the M&A surveyed. Section 7 concludes the chapter.
4.2. Literature review
4.2.1. The misvaluation hypothesis
Shleifer and Vishny (2003) developed a theoretical model of M&As based on the assumption that the
market is inefficient whereas managers are rational. With the assumption of the misvaluation
hypothesis, managers are able to explore the market’s mispricing given that firms are not fairly valued.
The authors suggest relative valuations of the two firms involved in an M&A deal as an important
factor explaining why M&As are initiated, the method of payment choice for financing M&As, and
the valuation consequences of the M&A wave. Their theoretical model predicts that bidders sacrifice
the short-term benefits in exchange for the firm’s long-term value. Overall, the model has raised two
important implications to be investigated: the method of payment choice and the long-term
performance of the combined firm.
Chapter 4. Relative Reference Prices and M&A Misvaluations
172
The misvaluation hypothesis is related to the method of payment hypothesis. It suggests that the
method of payment choices would reflect bidders’ M&A motive. Bidders who offer targets with
stocks have long-term horizons, in that they aim to create value instead of reaping mispricing benefits
in the short run. Shleifer and Vishny (2003) suggest that the net long-term effect of stock deals would
generate short-term losses but long-term ‘gains’. Though stock-financed acquisitions bring forth
negative returns to the bidder in the long run, those acquisitions benefit bidders who pay with stocks
that were significantly overvalued prior to any acquisitions. As noted by Shleifer and Vishny (2003:
301), ‘returns are just not as negative as they would have been without the acquisition’, which
suggests that overvalued stocks used as a means of payment for acquisitions are treated as a cushion
for the collapse of the stocks in the long run. In this respect, a firm’s value is enhanced with
overvalued stocks even though there are negative long-term returns.
Moreover, the misvaluation hypothesis suggests that the method of payment clarifies the market
important information about a firm’s value. It is generally believed that bidders who pay targets with
stocks would signal to the market that their firms are overvalued or their firms are relatively more
overvalued than the targets, whilst targets paid with cash are perceived to be undervalued relative to
their fundamental value. This is also consistent with the signalling hypothesis of Myers and Majluf
(1984). The major difference between the two hypotheses regarding the method of payment is that
the signaling hypothesis assumes that investors are asymmetrically informed in that they do not have
private information about the other firm and pay with stocks to deal with uncertainty, whereas the
misvaluation hypothesis relaxes this assumption and posits that the primary motive of paying with
stocks is that managers try to time the market and aim to create value for the firm. Therefore, the
misvaluation hypothesis interprets the long-run reason for the bidder to finance an M&A deal,
Chapter 4. Relative Reference Prices and M&A Misvaluations
173
reasoning that managers who pay with overvalued stocks try to time the market.
The theoretical model of misvaluation hypothesis has given rise to many important implications. First,
the issue on whether all-stock acquisitions are driven by misvaluation has raised doubts. Savor and
Lu (2009) imply that all stock acquisitions are driven by misvaluation, assuming that managers try to
time the market by paying for targets with overvalued stocks and highly valued bidders have the
greatest incentive to undertake an acquisition. Their study focuses on long-term performance of stock
bidders and finds that though stock acquisitions destroy the firm value, stock bidders of successful
deals outperform those who fail to complete an M&A deal. Analysing a sample of 12,578 U.S.
mergers between 1962 and 2000, the authors found that stock acquisitions generate long-term
negative abnormal returns for a firm. Despite this, the mean difference of buy-and-hold abnormal
returns between successful deals and unsuccessful deals increase from 13.6% to 31.2% over the three
years following an acquisition,55 suggesting that stock acquisitions are a result of managers timing
the market. In contrast, Fu et al. (2013) find evidence that not all stock acquisitions are driven by
misvaluation. Among 1,319 stock-financed acquisitions studied in their sample, one third is not
motivated by stock overvaluation, which is neither relative to its fundamental value nor relative to
the target valuation. Eckbo et al. (2016) documented more direct evidence contradicting the notion
that stock acquisitions are market-driven. Investigating a sample of 4,919 merger bids between 1980
and 2008, the authors found that bidders paying with stocks as a means of payment for acquisitions
tend to be small firms and with low leverage, implying that the reason for bidders to pay with stocks
is that the firm is financially constrained and limited in free cash flows. In addition, stock acquisitions
55 BHARs were calculated with the market-adjusted model. Their paper shows a similar trend when BHARs were
calculated with the calendar-time portfolio model.
Chapter 4. Relative Reference Prices and M&A Misvaluations
174
are more likely to occur in high-tech industries and when target and bidder are in geographical
proximity, suggesting the reason for targets to accept stocks is because that they have information
about the true position of the bidder, eliminating the concern of information asymmetry.
The second concern relating to the misvaluation hypothesis is whether bidders paying with
overvalued stocks could create value for the firm. Shleifer and Vishny (2003) suggest that the long-
run reason for the bidders to use stocks is to create value for the firm as holding stocks in an
overvalued market will collapse the firm’s value when the market incorporates information and
downgrades the value accordingly. With this in mind, bidders are rational. Consistent with this
prediction, Ang and Cheng (2006) found that the long-term performance of stock bidders and
overvaluation are positively correlated. Analysing a sample of over 3,000 mergers between 1981 and
2001, the authors found significantly positive abnormal returns for stock bidders in the estimation
period of one day prior to the takeover announcement date and three years following merger
completion. The results continue to hold compared with those drawn from the control group that
contains similarly overvalued firms who do not undertake an acquisition. The result shows that
overvalued bidders outperform those similarly overvalued non-acquirers.
Lin et al. (2011) explain underperformance as long-run reversal of initial overvaluation rather than
an overpayment. Their quartile analysis results of buy-and-hold abnormal return by bidder’ P/V show
that overvalued bidders underperform their undervalued counterparts in the long run over three years
following an acquisition. Their results are robust in the situation where the offer premium and other
relevant variables were taken into consideration in the multivariate regression of BHARs on bidder
P/V. The rationale of controlling for the offer premium is that bidders with high P/V are also those
Chapter 4. Relative Reference Prices and M&A Misvaluations
175
paying high M&A offer premiums, leading to the same prediction that overpayment results in
underperformance. In order to capture the net effect of the market correction on the firm overvaluation,
the authors used multivariate regressions to control for the offer premium after univariate analysis. In
addition, their data show that the average offer premium for the most overvalued bidders are
insignificantly different from the least overvalued bidders, 56.21% and 53.64% respectively, thus
weakening evidence that overpayment triggers underperformance in the long run.
Fu et al. (2012) argued that long-term underperformance is due to overpayment. They found that
overvalued firms tend to pay too much for acquisitions and attribute this to the weakness of the
corporate governance structure of a firm. The authors measure post-merger performance of the firm
with both the market-based approach as well as the accounting-based approach. Using the accounting-
based approach is based on the rationale that the initial overvaluation of the firm will be corrected
regardless of the occurrence of an acquisition. This may not be captured by conventional market-
based approaches. The authors thus assessed the firm’s long run performance with the firm’s ROA,
which is a proxy for firm’s operational performance rather than the market reaction, taking managerial
effectiveness into account. Overall, both approaches indicate negative long-term synergies for the
firms whose corporate governance structure is of inferior quality.
Since the long-term benefits of an acquisition are nearly zero as predicted by the misvaluation
model,56 leading to a question as to why target managers are willing to accept bidders’ overvalued
stocks. Shleifer and Vishny (2003) interpreted this as target’s short-horizons. With this in mind,
targets tend to voluntarily accept overvalued stocks to raise the liquidity of the firm or simply trade
56 Given the fact that value creation for the bidder, in return, damages the target in the long run.
Chapter 4. Relative Reference Prices and M&A Misvaluations
176
them after acquisitions for a cash-out purpose. Another explanation is that this matter is attributed to
the agency problems of the target firms where managers are offered with lucrative benefits, such as
good pay or being retained in the top position of the bidder firm following acquisitions, leading them
to give up control of the firm.
However, the assumption of short-horizon targets has been questioned by a number of M&A studies.
Rhodes-Kropf and Viswanathan (2004) suggest a rational target model whereby rational targets
should not be willing to accept overvalued stocks, as they clearly know about the danger of holding
stocks in an overvalued market. The authors suggest that both target and bidder managers are well-
informed about their own firms but they lack private information about the other firm. Such private
information of the firm tells managers whether the other firm is undervalued or overvalued but cannot
tell managers what drives such misvaluation. A rational target is able to assess the synergies of the
merger and decide not to accept overvalued stocks so as to prevent the value of the firm from being
damaged. However, the view is distorted during the periods when market-wide valuation is high. As
it is explained by Rhodes-Kropf and Viswanathan (2004), ‘the rational target correctly filters on
average but underestimates the market wide effect when the market is overvalued and over estimates
the effect when the market is undervalued.’ Thus, an overestimation of synergy positively correlates
with valuation errors. Despite target managers being rational, they may find it hard to distinguish
synergies from misvaluation without knowing what drives misvaluation. Their data indicate that
stock-financed acquisitions are prevalent during high market-wide valuation periods when rational
targets tend to make mistakes by wrongly accepting bidders’ overvalued stocks.
Following this, Rhodes-Kropf et al. (2005) suggest that rational targets tend to accept overvalued
Chapter 4. Relative Reference Prices and M&A Misvaluations
177
stocks when valuation errors mislead their estimation on synergies. The authors find the volume of
stock acquisitions and valuation errors are positively correlated. These authors use regression
techniques to break the market-to-book ratio (M/B), the conventional misvaluation measure, into
three individual components: the firm-specific misvaluation, the sector-wide misvaluation and the
long-run value to book which captures long-run growth opportunities. Their data suggest that the
firm-specific level and the sector-level are the most important factors driving misvaluation. More
specifically, bidders on average have higher firm-specific errors than targets, which is on average
60%. Misvaluation can reconcile about 15% of merger activity at the industry level. Rhodes-Kropf et
al.’s study (2005) shows that low long-run value-to-book firms buy high long-run value-to-book firms,
suggesting that low growth opportunities firms acquire high growth opportunities firms. Overall,
misvaluation-driven mergers are as a result of valuation errors, contradicting to what the misvaluation
hypothesis assumes.
Likewise, Di Guili (2013) finds rational targets, suggesting that target managers care about the quality
of mergers rather than private benefits. In this regard, stocks are associated with better investment
opportunities. So as to distinguish the effects of misvaluation and growth opportunities, the author
proposes a new proxy for investment opportunities of the firm and finds there is positive correlation
between stock payment and the firm’s prospects. Following this line of literature, Vermaelen and Xu
(2014) explained rational targets with capital structure theory. They found that rational target
managers are likely to accept overvalued stocks as they believe bidders paying with stocks will reach
an optimal capital structure. They proposed a prediction model regarding the method of payment,
which predicts that the bidder pays cash when the market expects cash payment. If bidders pay cash
when the model predicts stocks, the long-term abnormal returns will be positive based on the rationale
Chapter 4. Relative Reference Prices and M&A Misvaluations
178
that the market under-reacts to the firm. Whilst bidders pay with stocks as the market predicted, their
long-term abnormal returns will be negative, assuming that the market can identify the motive for
financing an M&A is overvaluation. The prediction model explains 89% of Vermaelen and Xu’s
M&A sample (2014). It indicates that cash can generate positive long-term abnormal returns on the
condition that the market recommends stocks, whereas bidders should pay cash when the model
predicts cash, indicating that target managers are rational in that they would not accept overvalued
stocks without a reasonable explanation.
The misvaluation hypothesis explains who buys whom and the medium of payment. Dong et al. (2006)
conducted a thoroughly investigation into the relationship to these predictions. The authors analysed
pre-takeover valuation of targets and bidders by using the firms’ market price-to-book value of equity
(P/B), and market price-to-residual income (P/V). This latter misvaluation measure has been largely
ignored by previous studies. Their sample consists of 3,732 U.S. acquisitions announced between
1978 and 2000. Their study suggests that these two misvaluation measures are highly correlated but
P/B tends to have more explanatory power over the misvaluation hypothesis than P/V on the rationale
that managers are becoming more familiar with P/B than P/V. According to Dong et al. (2006), P
denotes the market value of the firm while both B and V denote the firm’s fundamental value. Thus,
the essence of these two misvaluation proxies indicates the extent to which a firm’s market value
deviates from its fundamental value. The difference between B and V is that, the former measures the
firm’s book value whilst the latter contains analysts’ forecasting information. Therefore, P/B is a
proxy for the firms’ growth opportunities or managerial effectiveness, which is used to investigate
the Q hypothesis, whilst P/V is more likely a pure proxy for misvaluation in this context given that
the measure reflects expectations of analysts’ forecasts about the firm’s future performance. However,
Chapter 4. Relative Reference Prices and M&A Misvaluations
179
it is argued that analysts’ forecasts may not accurately reflect the information about the firm’s
prospects and managerial effectiveness. Therefore, the authors propose to include both P/B and P/V
to investigate the misvaluation hypothesis.
Dong et al.’s study (2006) focuses on the relationship between the valuations of both the bidder and
the target and many aspects of M&As, including the method of payment, M&A offer premiums, M&A
outcomes and a wide range of deal characteristics. The sample includes 2,922 successful and 810
unsuccessful takeover bids announced between 1978 and 2000, and filters out those M&A deals
whose transaction value was less than $10 millions, eliminating concerns about small deals
confounding the result.57 Dong et al.’s data show that bidders are generally those with a higher
valuation whereas targets are those with a lower valuation, reflected in higher P/B and P/V of bidders
opposed to targets. Once again, a higher valuation firm is more likely be the takeover bidder than a
lower valuation firm, which is consistent with the prediction in Shleifer and Vishny’s theoretical
model (2003). Stock-financed takeover bids show significantly higher valuation of the bidder and the
target than cash-financed takeover bids. Of particular note, the mean valuation differential between
the bidder and the target are highly significant in the two misvaluation measures in the stock-financed
acquisition subsample, whilst P/B has more explanatory power than P/V in explaining the difference
in bidder-target valuation in the cash-financed acquisition subsample. With respect to M&A outcomes,
the authors noted that bidders with higher P/B tend to pay targets with higher M&A offer premiums,
which are associated with negative market reactions to the bidder and positive market reactions to the
target.
57 The authors also winsorised P/B and P/V of the firms at 1% and 99% levels for a similar reason.
Chapter 4. Relative Reference Prices and M&A Misvaluations
180
Their study proposes two different proxies for misvaluation and find some rather mixed results when
relating to a wide range of takeover characteristics. The authors found that cash bidders are also
relatively more overvalued than targets. The misvaluation effect triggers bidder managers’ incentives
of exploring mispricing targets. As a result, rather than paying targets with stocks, bidders who are
more overvalued would also pay cash to the undervalued targets. This argument complements stock-
market driven acquisitions, reasoning that either cash or stocks may be used when the bidders’ M&A
motive is to dilute overvaluation. The authors also note that these proxies for misvaluation are a
reflection of the managers’ perceptions of a firm’s valuation rather than the true position of the firms,
explaining relative valuations from a behavioural finance perspective. Despite that, the study has
thoroughly investigated the misvaluation hypothesis and provides empirical evidence relating to what
the theoretical model of misvaluation had proposed, failing to provide any rationale as to why target
managers accept overvalued stocks. It remains unclear whether target managers accept bidders’ stocks
believing that bidders with high P/B or P/V are those with good past performance and thus are able
to create synergies for the combined firm.
With this in mind, Rhodes-Kropf and Viswanathan (2004) and Rhodes-Kropf et al. (2005) proposed
that targets accept overvalued stocks is because they are misled by valuation errors. Whilst Burch et
al. (2012) suggest that targets prefer to hold highly valued stocks as they believe those firms are
normally well-run and have better prospects, implying that targets can identify the sign of
overvaluation and voluntarily accept those highly valued stocks. Their findings are consistent with
the prediction of the Q hypothesis. By accepting those highly overvalued stocks, targets could
liquidate their holding position.
Chapter 4. Relative Reference Prices and M&A Misvaluations
181
Many additional studies have proposed various proxies for misvaluation, which challenge the
conventional misvaluation measures. For example, Ben-David et al. (2015), who used short interest
as a proxy for misvaluation, reflecting investors’ under- and over-reactions to the firm’s valuations.
The authors found that stock bidders have higher short interests relative to targets. The use of short
interest as a measure for misvaluation isolates the effect of growth opportunities which may lead to
the same prediction that stocks are likely to be used as a means of payment for M&As. Ben-David et
al.’s results are consistent with the misvaluation hypothesis. Di Guili (2013) challenges MTBV by
arguing that it represents both the misvaluation and growth opportunities of a firm. It is generally
believed that firms with a high MTBV may be a result of overvaluation or better investment
opportunities. So as to distinguish the effects of misvaluation and investment opportunities, the author
develops a new measure for investment opportunity, which is based on post-merger investment.
Analysing a sample of 1,187 mergers announced between 1990 and 2005 and using the average ratio
of capital expenditures over assets in the four years following the merger as a proxy for the firms’
investment opportunities, the author finds that the propensity for using stocks increases with the post-
merger investment ratio. His study indicates that the firm’s market value relative to its fundamental
value is a problematic proxy for misvaluation and likely to bias the predictions of the misvaluation
hypothesis.
4.2.2. Reference point theory of M&As
The reference point is an essential implication of Kahneman and Tversky’s prospect theory (1979).
It portrays the concept that people’ s feelings about the gain and the loss are judged relative to a
particular reference point, which is normally the current position of a subject, the status quo or the
aspiration level, as opposed to a reference point. Reference points distinguish prospect theory from
Chapter 4. Relative Reference Prices and M&A Misvaluations
182
the traditional expected utility theory that emphasise the final outcome of the wealth level but fails to
account for the decision-making process.
The reference point theory is rooted in Tversky and Kahneman’s anchoring and adjustment
mechanism (1974) suggesting that people tend to rely heavily on a piece of salient but perhaps largely
irrelevant information at an initial stage of the decision-making process and adjust the final estimates
when new information is incorporated. This psychological phenomenon refers to the belief formation
process. Many implications of prospect theory are largely built upon the reference point theory. There
exists a loss-aversion tendency, given an equal-size of loss or gain, the feelings about the loss are
more pronounced than those about the gain (Kahneman and Tversky, 1979), and there also exists a
diminishing sensitivity that the feelings about the gain and loss are more pronounced in the scenario
where the outcome is close rather than distant in relation to the reference point (Tversky and
Kahneman, 1992).
The reference point theory interprets how people make decisions under the situation where
considerable information uncertainty exists. Baker et al. (2012) were the first to relate this theory to
the M&A pricing decision. They noted that the use of the target 52-week high as the reference price
lies in two proposed reasons. First, the firm’s 52-week high is frequently reported by the financial
media, which reinforces the investors’ formation process to utility realisation. Secondly, Baker et al.
justify the rationale behind the use of the target 52-week high from the psychological and practical
points of views of both the target and the bidder firms.58
58 The researcher discussed the main findings of their paper in the previous Chapters.
Chapter 4. Relative Reference Prices and M&A Misvaluations
183
The authors also document evidence of the reference point effect on the market. When assessing the
deal outcome, the authors found that there are high offer premiums to targets who negotiate bidders
with their 52-week highs, whilst bidders receive negative market reactions, indicating that the offer
price driven by reference price is due to an overpayment. On the whole, Baker et al.’s study (2012)
suggests an evident reference point phenomenon for both the individual and the institutional investors.
Chira and Madura (2015) extended Baker et al.’s work (2012) by interpreting the role of reference
point prices as the bargaining power of the firm. Chira and Madura focused on the relationship
between takeover possibility and the reference points of the two firms involved. Similar Baker et al.’s
work (2012), Chira and Madura measured the reference point as the deviation of the firms’ 52-week
highest stock price from their current prices. The study assumes that the firms have greater bargaining
power when its current price is closer to, rather than distant from, its highest stock price over the year.
There results show that firms are likely to be takeover bidders when they have less distance between
their current price and the highest price over the year, indicating that a bidder’s great bargaining
power leads to a greater incentive of undertaking acquisitions. Moreover, it was found that firms are
less attractive to unaffiliated firms when the distance between the reference price and the current price
is large, arguing that the firm whose price deviates considerably from its 52-week high tends to resist
the deal unless a high price is offered in compensation. And because of this, bidders will abandon the
deal.
The authors suggest that both the target and the bidder will assess the firm value with their own
reference points. By examining the relationship between deal completion and the firms’ reference
points, the authors argue that targets with a low stock price relative to their 52-week highs may
Chapter 4. Relative Reference Prices and M&A Misvaluations
184
perceive themselves as undervalued, resulting in strong deal resistance. Bidders will pursue the deal
by paying for takeover targets with stocks whose price is high relative to their 52-week highs, given
that bidders perceive their firms to be overvalued. However, their study fails to provide any rationale
on why bidders would look at their own reference points since they have more information about their
firm than other investors in the market. In addition, the study argues that the deal is likely to be
completed when the price distance between the firm’s 52-week high and the current price is small for
both the target and the bidder, relying on the explanation that psychological barriers between the two
firms are low.
Many additional studies have indicated an important role the firm’s reference point price plays on
investors’ trading strategies. George and Hwang (2004) used a firm’s 52-week high as the reference
point price and found evidence of reference dependence bias in the stock market. They note that
investors view the price that is close to its 52-week high price as good news and the price that has a
great distance its 52-week high as bad news. The mechanism is that investors are unwilling to bid up
a price when it is close to the 52-week high and should be reluctant to press down a price when it is
a long way below the 52-week high. This reasoning suggests that investors have a tendency to make
decisions based on a reference point price, which is the firm’s 52-week high. The salient evidence of
reference dependence is also supported by the findings of Grinblatt and Kelaharju (2001) who
investigated the Finnish stock market. They used a firm’s historical high as a reference point price
and found that investors avoid selling (buying) stocks whose value is far below (close) to the historical
high. Huddart et al. (2009) observed the phenomenon of abnormal sales’ volume around the 52-week
high based on a sample of 2,000 firms over 24 years.
Chapter 4. Relative Reference Prices and M&A Misvaluations
185
Shefrin and Statman (1985) put forward the disposition effect that investors tend to sell winner stocks
too soon and hold loser stock too long, which is in favour of prospect theory. Barberies and Xiong
(2009) also documented evidence of loss-aversion based on a large sample of activities by individual
investors. These authors show investors’ greater propensity for selling stocks whose value has risen
recently and buying those whose value has been downgraded recently. According to a survey of
reference point theory (Olsen, 1997), it often occurs that the reference point effect is evident among
both individual and institutional investors, which implies that reference dependence bias is human
nature when facing uncertainty.
Bidder and target managers look at the reference point of the firms because they know the market
also has a reference-dependence bias. In this case, bidder managers can justify the motive for M&A
offer premiums paid for the target. Targets demand offer premiums around the target 52-week high
to avoid being blamed for selling the firm with a low valuation. Though Baker et al. (2012) showed
that offer premiums are positively related to the target 52-week high, it is less likely for the researcher
to know the whole reason for the bidders’ M&A motive without considerations about the bidder’s
side. For example, bidders are overconfident as they would argue that paying for targets based on
their 52-week highs can create synergies. It may also be the case that bidders are in a relatively weaker
bargaining position to targets because of information asymmetry, indicating that offer premiums
relative to the target 52-week high are due to target’s strong bargaining power. This chapter puts
forward Baker et al.’s idea (2012) by adding the role of the bidder reference point, and explores how
M&A activity is structured, based on the reference point of the two firms involved.
Chapter 4. Relative Reference Prices and M&A Misvaluations
186
4.3. Hypothesis development
Shleifer and Vishny's misvaluation hypothesis (2003) indicates that bidders have a direct incentive to
use stocks as means of payment to finance an M&A deal when they are overvalued. It is the case that
overvaluation is a managerial perception as noted by Dong et al. (2006). The market has a tendency
to assess the firm value based on the reference point price, as predicted by reference point theory of
M&A (Baker et al., 2012). Managers whose aim is to time the market will also consider this and
eliminate such overvaluation concerns through M&As. Bidders are likely to be those with a stable
performance relative to targets, drawing its current price closer to the 52-week high, rather than the
case of the target. This gives rise three possible situations when the firm’s value has recently moved
toward to its 52-week high. If bidders are fairly estimated, nearness to the 52-week high price
indicates bidder’s strong bargaining power. The market believes that the firm is strong in making
profits for the combined firm, and as a result, bidders would pay fewer offer premiums. The second
scenario is that bidders are possibly undervalued even though there is nearness to the 52-week high
price given that the firm’s growth opportunities may be underestimated. The arrival of the information
at a later stage will push the firm’s price to a new 52-week high. If so, bidders would possibly pay
for targets with cash rather than undervalued stocks. The third scenario is that bidders are overvalued
with nearness to the 52-week high price. If it is the case, bidders are likely to pay for a target with
stocks, aiming at diluting overvaluation and creating firm value in the long run. Furthermore,
eliminating overvaluation requires additional costs, which result in a high offer price in excess of the
target’s current value.
The RRP eliminates any of these concerns on the bidders’ value, as it reflects the extent to which the
bidder is relatively more overvalued than the target from the market perspective. This proxy
Chapter 4. Relative Reference Prices and M&A Misvaluations
187
investigates the valuation at the market level rather than at the firm level, which is the misperception
of the firm’s valuation created by those investors lacking private information about the firm. If
managers can time the market, M&As are a valuation game. It should be suggested that bidders should
dilute their valuations as long as they are relatively more overvalued than the target. One important
sign of this overvaluation is that bidders choose stocks as a means of payment for finance M&As,
since holding stocks in an overvalued market will destroy the wealth of long-term shareholders.
Therefore, the probability of using stocks for payment purposes increases in line with movements in
the RRP leading to our first testable hypothesis of
H1: There is a positive correlation between the RRP and the likelihood of using stocks as a means of
payment in M&As.
Shleifer and Vishny (2003) predicted that bidders’ overvaluation will be diluted through acquiring an
undervalued (or less overvalued) firm. Dong et al. (2006) proposed P/B and P/V as proxies for
misvaluation, which is the firm’s market value relative to its fundamental value. Dong et al.’s findings
imply that diluting overvaluation remains a priority objective for bidder managers regardless of the
method of payment used for finance M&As. It might be the case that cash bidders are also overvalued,
in that overvaluation drives acquisitions. According to this argument, bidders make a takeover bid
with cash rather than not bidding at all. Thus, bidders can profit from acquisitions by using stocks
when they are relatively more overvalued and using cash when targets are undervalued. This
reasoning implies that M&As result from bidder managers timing the market.
Relative bidder-target valuations were measured with RRP, capturing the market perception of a
Chapter 4. Relative Reference Prices and M&A Misvaluations
188
firm’s valuation and also giving consideration of the valuation of the two firms involved. Since both
target and bidder managers will assess their firms’ value according to the reference point (Chira and
Madura, 2015), they can identify the signs of relative valuations. Thus, bidders would pay offer
premiums in exchange for the control of the target firm. Targets would demand high offer premiums
based on this relative valuation, as they can also identify any misvaluation reflected in the RRP. They
would resist the deal unless high offer premiums were offered in compensation. Based on the above
argument, it was tested that
H2: M&A offer premiums correlates positively with the RRP.
Finally, bidders with a relatively smaller changes related to the firm’s 52-week high (with a price that
is close to its 52-week high) are expected to have a stronger bargaining position, leading to low offer
premiums being paid to the target. Thus, a high offer premium would offer shareholders a
straightforward sense that there is an overpayment, thus reacting negatively to the bid announcement.
On the other hand, this researcher argues that M&As serve as a value enhancement opportunity for
the target firm as shareholders would expect that targets with a relatively higher reference point would
have a higher probability of profiting through acquisitions. Therefore, our last testable hypothesis is
H3: Bidder (Target) short-term performance correlates negatively (positively) with the RRP.
4.4. Data and methodology
4.4.1. Data
The initial sample covers 36,506 U.S. domestic public acquisitions announced between January 1,
Chapter 4. Relative Reference Prices and M&A Misvaluations
189
1985, and December 31, 2014, as provided by Thomson One. Stock price was collected from CRSP,
and a series of standard accounting variables were collected from COMPUSTAT. Accounting
variables were required to be available for the fiscal year end prior to the announcement date are
required. Public acquisitions refer to the two firms involved being publicly traded U.S. firms (listed
on NYSE/AMEX/NASDAQ).59 Once deals that were classified as recapitalizations, repurchases,
self-tender offers and rumors according to Thomson One were excluded, it was left with 11,615
observations.60 The offer premium is not a missing value, which further reduces the number of
observations to 5,450. The stock price for the calculation of the bidder and the target 3-day CARs
were required to be available, which left with 4,630 observations. The payment method information
to be available in Thomson One, which left the chapter with 4,290 observations. It yielded 2,156
observations after excluding all bidder variables with a missing value, and has a final sample of 1,878
observations after excluding all target variables with a missing value.61
The RRP effect on the probability of using stocks as a means of payment for acquisitions was analysed
by controlling for a series of deal, bidder and target characteristics, which are standard control
variables highlighted in M&A literature. Bidder size is expected to be negatively related to the stock-
financed acquisition. Faccio and Masulis (2005) suggest that larger bidders have higher credit
facilities, which reduces the probability for using stocks. The firm’s growth opportunities were
measured with MTBV. Higher MTBV bidders tend to use more stocks in acquisitions, in that they
59 Firms should have an available stock price to calculate the 52-week high.
60 By doing these, acquisitions in which the bidder and the target is the same firm were eliminated, on the rationale that
the difference between the target and the bidder reference points cannot be distinguished (e.g. self-tender offers), and
acquisitions in which the target actively searches for the bidder were also excluded, since those deals violate the reference
point effect. 61 All variables that have a missing value and were used in the regressions were excluded, following a summary of the
acquisition sample and variables.
Chapter 4. Relative Reference Prices and M&A Misvaluations
190
reserve cash to fund new investment projects (Rhodes–Kropf et al., 2005; Dong et al., 2006). The
firm’s profitability was measured with the return-on-asset ratio (ROA). It should be suggested that
firms with higher profitability are more likely to use retained earnings held in cash rather than stocks
as it reduces costs of financing. Target characteristics were also accounted for, since stocks are more
likely to be used to mitigate the target risk (Hansen, 1987). In this respect, the propensity for using
stocks is greater when targets’ risks increase, such as large size, with a high MTBV and a low ROA.
Following Officer (2004), information asymmetry is measured by calculating the standard deviation
of returns. Hansen (1987) suggests that stocks are more likely to be used when level of information
uncertainty increases. Leverage is defined as debt-to-equity ratio (D/E). Vermaelen and Xu (2014)
suggest that over-levered bidders who justify stock financing in terms of moving to an optimal capital
structure lead to an increase of overvalued stocks for acquisitions. Whereas targets with high leverage
should be reluctant to receive such stocks. Liquidity is defined as cash flow-to-equity ratio (CF/E).
Higher liquidity firms are more likely to be less financially constrained firms, which result in the
method of payment for acquisitions is cash rather than stocks. Inclusions of capital structure related
variables in the regressions would allow us to disentangle the effects of firm’s capital structure
decision and misvaluation on stock-financed acquisitions.
In a further analysis, the RRP effect on the offer premium was analysed. Different categories of
variables were controlled in line with the work of Eckbo (2009). Specifically, firm size was measured
with logarithm of market valuation (MV). According to the hubris management hypothesis (Roll,
1986), larger bidders tend to pay generously for smaller targets. The firm’s profitability was measured
with ROA. Agency theory suggests that poor-performing bidders tend to dissipate firms’ resources
Chapter 4. Relative Reference Prices and M&A Misvaluations
191
and overpay for the target, whereas well-performing firms are attractive to bidders (Schwert, 2000).
The firm’s growth opportunities were measured with MTBV. Rau and Vermaelen (1998) suggest the
extrapolation hypothesis that glamour bidders are less cautionary than value bidders about the target
valuation, leading to higher offer premiums. Harford (1999) suggests that the target MTBV links with
the managerial takeover motive since bidders are more aggressive in exploring synergies from the
lower MTBV target. Stock volatility were calculated with standard deviation of returns over 335
calendar days ending 30 calendar days prior to the announcement date. All regressions include year
and industry effect.
The M&A sample was matched to firms’ characteristics collected from COMPUSTAT, and to stock
price information collected from CRSP. By doing this, all relevant information was presented in one
data file for the purpose of conducting analyses. MV is defined as the product of market price and
outstanding shares (CRSP: SHROUT*PRC). “Relative Size” is defined as the deal value divided by
bidder MV. ROA is return-on-asset ratio, defined as net income (Compustat: NI) divided by total asset
(Compustat: AT). MTBV is market-to-book value, defined as the market value of equity to the book
value of equity, where book value of equity is total shareholders’ equity (Compustat: SEQ) plus
deferred taxes and investment tax credit (Compustat: TXDITC) minus the preferred stock redemption
value (Compustat: PSTKRV). Capital structure related variables were also included in the regressions
of the probability for using stocks on RRP. Leverage was measured by debt-to-equity ratio, defined
as total long-term debt (Compustat: DITT) divided by the book value of equity. CF/E is cash flow-
to-equity ratio, defined as income before extraordinary items (Compustat: IBC) plus depreciation and
amortization (Compustat: DPC) minus cash dividends (Compustat: DV).
Chapter 4. Relative Reference Prices and M&A Misvaluations
192
4.4.2. Summary statistics
Table 4.1 reports a summary of the acquisition sample. Stock-financed acquisitions are those refers
to acquisitions that are 100% financed by stocks. Cash-financed acquisitions are those refers to
acquisitions that are 100% financed by cash. “Mix” refers to acquisitions that are neither all stock-
financed nor all cash-financed acquisitions. “Completed” refers to completed deals in the sample
period investigated in our chapter. “Tender” denotes tender offers, referring to deals that are proposed
directly to shareholders. This opposes to mergers, which made through negotiation of the
management teams of the two firms involved. “Hostile” denotes to hostile bids, portraying deal
attitude. “Diversification” refers to diversified deals in which the primary two Standard Industry
Classification codes are different between bidders and targets.
The average transaction value for 1,878 M&A deals in the sample is 1.54 billion US dollars. Of 1,878
acquisitions, 608 all stock-financed acquisitions, 726 all cash-financed acquisitions and 539 mixed
acquisitions.62 There were 702 diversifying acquisitions, as opposed to 1,176 relatedness deals, and
1,597 successful acquisitions compared with 277 unsuccessful deals.63 There were a small proportion
of tender offers and hostile acquisitions, 380 and 134 respectively.
Table 4.2 reports summary statistics for variables. Panel A presents the dependent variables used in
OLS regressions, including offer premiums, the bidder and the target 3-day CARs calculated using
the market model. The mean value for offer premiums is 31%. It is suggested that bidders have to pay
on average over 30% in excess of the target’s market price to gain the firm control. The mean value
62 The method of payment information for five acquisitions is defined as “Others” in Thomson One.
63 There were 1,874 deals with information about deal status based on Thomson One.
Chapter 4. Relative Reference Prices and M&A Misvaluations
193
for bidder CARs is lower than target CARs, 22.2% over -1.1%. Panel B presents the main variables
of interests. The mean value for the bidder reference point is lower than the target reference point,
29.4% to 41.2%, which suggests that bidders on average are relatively more overvalued. The mean
value for RRP is 11.8%. Panel C presents all control variables. Bidders are on average larger than
targets, 7.50 as opposed to 5.21. Bidders have better investment opportunities than targets, 3.99
compared to 2.69. The profitability for bidders is higher than for targets, 2.9% over -4.9%, reflected
in ROA. Overall, firms with larger size, higher profit-generating ability and better investment are
likely the takeover bidders whereas those with smaller size, lower profitability, and lack of investment
opportunities are likely the takeover targets. The summary statistics are generally consistent with
prior M&A literature (Fuller et al., 2002; Moeller et al., 2004).
4.4.3. Methodology
4.4.3.1. Relative reference point (RRP) and offer premiums
Reference point (RP) refers to what extent the current price deviates from its 52-week highest price.
The 52-week highest price is a highly relevant piece of price information that shapes investors’ minds
to the firm’s prospects. Lacking information and time availability to process the deal information, the
market would naturally borrow this to compare with the current price. Hence, the firm’s 52-week
highest price is the market reflection of the firm’s best performance. According to George and Hwang
(2004), the 52-week highest price is the outcome of a series of good news that occurred in the past
that drives the firm’s market value beyond its fundamental value. Thus, the 52-week high is largely
related to the market’s perception of the firm’s valuation. A lower reference point would indicate that
the firm is still in the momentum of the “good news” effect, leading firms to be more overvalued. In
contrast, a higher reference point indicates that the good news effect is less relevant, leading firms to
Chapter 4. Relative Reference Prices and M&A Misvaluations
194
be less overvalued. Therefore, a target reference point (TRP), larger relative to the bidder reference
point (BRP), sending a signal of market-wide perception regarding the target firm being less
overvalued than the bidder firm. Furthermore, the extent to which the bidder is more overvalued than
the target is measured with the relative reference point (RRP). Therefore, the market misperception
of firm’s valuation tends to reduce when a lower reference point firm mergers with a relatively higher
reference point firm. The data also show that TRP is on average larger than BRP.64 Formulas for RP,
RRP and offer premiums were illustrated as follows:
, 30 , 30log(52 ) log( )i i t i tRP WeekHigh StockPrice (4.1)
i i iRRP TRP BRP (4.2)
, , , 30log logi t i t i tOfferPremiums OfferPrice TStockPrice (4.3)
where iRP denotes the reference point of each firm i. The bidder (target) reference point is defined as
the logarithmic term difference between the bidder’s (target’s) highest stock price over 335 calendar
days ending 30 days prior to the announcement date and bidder’s (target’s) stock price 30 days prior
to the announcement date.65 iRRP denotes relative reference point which is defined as the target
reference point (iTRP ) and the bidder reference point (
iBRP ). Offer premiums were calculated as the
64 The difference between the target reference point and the bidder reference were used, attempting to obtain a positive
value, making the coefficients easier to interpret. According to the prediction, a less overvalued target should have a
higher reference point whereas a more overvalued bidder should have a lower reference point. The descriptive statistics
also show that, on average, the target reference point is larger than the bidder reference point: 0.412 and 0.294.
65 Using the firm’s stock price 30 days prior to the announcement date tends to mitigate the market reactions released due
to information leakage.
Chapter 4. Relative Reference Prices and M&A Misvaluations
195
logarithmic term difference between the offer price ( ,i tOfferPrice ) and target stock price 30 days prior
to the announcement date ( , 30i tTStockPrice ).
As this chapter examines the effect of a relative valuation between the two firms involved on an
M&As transaction but not the extent to which the firm’s market value deviates from its true value,
the use of the RRP allows the researcher to establish a bilateral valuation framework where it is the
valuation deviation of the two firms influences the market perception. The market-wide valuation
would potentially drive the firm’s market value away from its fundamental value (Rhodes-Kropf and
Viswanathan, 2004). By doing so, it can account for the influence of market swings, as the RRP is a
relative valuation measure allowing the researcher to observe a new equilibrium of the firms’
valuation.
4.4.3.2. Logistic regressions
Logistic model refers to that dependent variable is categorical. The dependent variable of binominal
logistic model takes two values. This model was employed to study whether acquisitions are purely
financed by stocks or other means of finance, taking value of 1 and 0 respectively. While the
dependent variable of multinomial logistic model takes more than two values. The model was
employed to study the magnitude change in the RRP causes the likelihood of the acquisitions are
financed by stocks rather than cash or the mix payment. Both binomial and multinomial logistic
models were employed in this chapter. Binominal logistic regression was used to examine the RRP
effect on the probability of using stocks. The logistic coefficients were transferred into marginal effect
(ME) at sample means, which provide consistent interpretation of OLS estimates. In addition,
multinomial logistic regression was used to examine the RRP effect on the probability of using stocks
Chapter 4. Relative Reference Prices and M&A Misvaluations
196
rather than other means of payment as a robustness check. The model was specified as follows:
, 1 , ,
1
( ) ( )N
i t i i i t i t
i
P Stock Pl RRP X
(4.4)
, , 1 , ,
1
( ) 1 ( ) ( )N
i t i t i i i t i t
i
Q Stock P Stock Pl RRP X
(4.5)
where Pl represents the possibility function, the main variable of interest to be investigated is iRRP .
iX denotes control variables in relation to the payment choices for M&A deals. (4.4) and (4.5)
combined yield the behavior of the logistic function, as follows:
'
1 , ,
1
( )N
i i i t i t
i
Pl RRP X
= ,( )i tP Stock ,( )i tQ Stock (4.6)
The function represents the derivative in respect of the ,i tRRP .
4.4.3.3. Classification of high-, neutral- and low-valuation markets
Rhodes-Kropf and Viswanathan (2004) suggest stock-financed acquisitions are positively correlated
with market-wide valuation. Following Bouwman et al.'s approach (2009), market valuation periods
were classified using the price-earnings (P/E) ratio of the market index (S&P 500) and monthly data.
Firstly, the market P/E ratio was de-trended by removing the best straight line fit (OLS) from the P/E
of the month in question and the five preceding years. Secondly, each calendar month was classified
into high- (low-) market valuation groups if the de-trended market P/E ratio of that month was above
Chapter 4. Relative Reference Prices and M&A Misvaluations
197
(below) the five-year average. Then, the months were ranked according to the de-trended market P/E
ratio. Months in the top 25% of the above average group were classified as high-market valuation
months, months in the bottom 25% of the below average group were classified as low-market
valuation months, the remaining months being classified as neutral-market valuation months. Thus,
half of the months were classified as neutral-market valuation and the other half contains months of
both high- and low-market valuation. The idea of de-trending market valuation is to remove the
upwards trend because the most recent acquisitions generally have a higher market valuation than the
past due to market inflation.
4.4.3.4. Short-term method
Following Eckbo et al.'s market model (2016), firms’ announcement returns were calculated as
follows:
, 1 , ,i t m t i tR R (4.7)
where ,i tR denotes holding period returns (CRSP: RET) for firm i in the period t, ,m tR denotes value-
weighted market returns including dividends (CRSP: VWRETD), ,i t denotes the error term. We
estimate market model parameters over the window from 261 to 28 trading days prior to the
announcement date [-261, -28], and use a 3-day event window [-1, 1].
4.4.3.5. Long-term method
4.4.3.5.1. Market-adjusted model
Following Loughran and Vijh (1997), the firm’s long-term performance was calculated with the
Chapter 4. Relative Reference Prices and M&A Misvaluations
198
market-adjusted buy-and-hold abnormal returns (BHARs). And 36-month BHARs were presented
with the following equation:
, , ,
1 1
(1 ) (1 )T T
i t i t index t
t t
BHAR R R
(4.8)
where, ,i tR is the arithmetic returns for firm i on day and ,index tR is the arithmetic return for the market
index on day t.
4.4.3.5.2. Size-adjusted model
Fama (1998) claims that the use of different long-term methodological approaches can generate
different results. He suggests the efficient market hypothesis that financial anomalies are chance
results due to the market misinterpretation of the information, such as under or over-reactions. They
are largely contributed to the bad-model problems. Fama (1998) suggests that the long-term return
anomalies tend to disappear with reasonable change in a way they measured. In the light of this, the
size-adjusted model was employed to assess bidders’ long-term abnormal returns. Lyon et al. (1999)
suggest that the use of this methodological approach can yield well-specific test statistics as it
alleviates the new listing and rebalancing biases. Following Lyon et al’s work (1999), BHARs were
measured by constructing reference portfolios based on all firms listed in NYSE. Specifically, size-
adjusted ARs were calculated as follows:
, ,
0
(1 ) 1T
i t i t pt
t
BHAR R R
(4.9)
Chapter 4. Relative Reference Prices and M&A Misvaluations
199
where ptR relates to reference portfolio return and calculated as follows:
,
0
1
(1 ) 1T
j tnt
pt
j
R
Rn
(4.10)
where ,j tR relates to simple return on the firm j at time t, and n denotes to the number of firms.
4.4.3.5.3. Bootstrapped t-statistics
Fama (1998) challenges the statistics for the univariate test of BHARs under conventional
methodological approaches. He argues positive-skewness problems that may not yield well-specified
t-statistics. In the light of this, a bootstrapped t-statistic and the non-parametric Wilcoxon rank-sum
test were used to estimate the differentials for BHAR ranks. The skewness-adjusted t-test is presented
as follows:
21 1
3 6T N S S
N
(4.11)
where
1, 2( )
BHAR
BHAR T TS
(4.12)
and
Chapter 4. Relative Reference Prices and M&A Misvaluations
200
1, 21, 2
3
( )( )
2
3
n
T Ti T T
i
BHAR
BHAR BHAR
N
(4.13)
where BHAR relates to the standard deviation of BHAR by conventional models.
4.4.3.6. Multivariate analysis
Multivariate analysis is employed to examine the relevant factors explaining the market reaction.
According to Draper and Paudyal (2008), multivariate analysis is superior in analyzing the causation
relationship between the market reaction and related variables. The multivariate framework is
presented as follows:
( 1, 1)
1
N
i i i
i
CAR X
(4.14)
where iX denotes variables related to market reactions. The main variable of interest to be
investigated is RRP, which is the difference between the target and bidder reference points. The
multivariate framework controls for a series of deal and firm characteristics that have significant
impacts on market reactions in standard M&A literature (Alexandridis et al., 2010; Asquith et al.,
1983).
4.5. Empirical results
In this section, empirical results are reported and discussions being provided in relation to the RRP
effect on the method of payment, offer premiums and bidder performance in the short and long runs.
Chapter 4. Relative Reference Prices and M&A Misvaluations
201
According to misvaluation hypothesis, overvalued bidders tend to time the market by paying
significantly high offer premiums to target shareholders and utilising their overvalued stocks. Despite
wealth being destroyed in the short run, the hypothesis predicts that rational bidders expecting to
create firm value in the long run. Following these predictions, logistic regressions of stock-financed
acquisition on the RRP were first conducted. Second, how offer premiums change according to RRP
is investigated. Our third main regressions examined the role of offer premiums plays on firms’
announcement returns. Specifically, the 2SLS technique is used to examine how announcement
returns of a firm change according to the RRP driven offer premiums. Finally, the post-merger period
performance of stock bidders is examined with univariate analysis by RRP ranking.
4.5.1. The RRP effect on the probability of using stocks
Table 4.3 reports a positive relationship between the RRP and the likelihood of using stocks for
financing M&As. The sign and significance level of the RRP do not change significantly after
information asymmetry has been included, as well as capital structure related variables of both the
bidder and the target.66 Specification (4) of Table 4.3 shows the main results, a one unit increase in
the RRP lead to about 10.4 percentage points increase of stock payments (p = 0.001), as interpreted
by marginal effect. In addition, an increase of MTBV results in an increase in the likelihood of the
use of stocks, suggesting that bidders tend to pay with stocks when they have better investment
opportunities on the reasoning that cash is likely to be retained for future investment. This is
consistent with Dong et al.’s view (2006) and that of Di Guili (2013). The results obtained show that
bidders who are small, with lower profitability and high debt are likely to pay with stocks, as they are
generally financially constrained which limits their cash borrowing. This is consistent with Eckbo et
66 Our results do not change significantly after considering the firm-fixed effect.
Chapter 4. Relative Reference Prices and M&A Misvaluations
202
al.'s findings (2016).
The main results suggest that the relatively more overvalued bidders are likely to pay with stocks,
which is consistent with the misvaluation hypothesis and also reconciles the predictions of the
reference point theory of M&As. Bidders whose price is close to their 52-week high would give
targets a chance of selling out overvalued shares for profits following acquisitions, which is consistent
with Burch et al.’s view (2012) that targets tend to accept overvalued stocks because that bidders can
generate profits in the future. Alternatively, targets would demand more stocks from bidders when
targets can identify overvaluation, which is in line with Vijh and Yang’s view (2014). When the
market news has driven the target firm’s current value away from its fundamentals, bidders have the
incentive to exploit the target price discounts. With the reference-dependence bias, bidders also assess
the M&A valuation by focusing on the RRP, arguing that if a target’s price falls greatly below its 52-
week high than that of their firms, would offer bidders greater potential for overvaluation dilution.67
Table 4.4 reports the RRP effect on the probability of using stocks under different market conditions.
Rhodes-Kropf and Viswanathan (2004) suggest that stock-financed acquisitions are positively
correlated with high market-wide valuation. This is because that the overestimation of synergies
increases with valuation errors. Consistent with this view, it is found that the RRP effect is more
pronounced when market valuation errors are sizable. As explained by the marginal effect in Table
4.4, for every one unit increase in the RRP would lead to an increase of about 15.3 percentage points
67 It can be argued that when the target current price is significantly lower than its 52-week high, the target may experience
risks of bankruptcy. It is believed that bidders would be cautious about this and may not focus on the target 52-week high.
However, these individual cases were not considered in this thesis, as there is only limited number of deals in this M&A
sample.
Chapter 4. Relative Reference Prices and M&A Misvaluations
203
when the market-wide valuation is high and an increase of about 10.9 percentage points when the
market-wide valuation is neutral, whereas the RRP effect on market condition is insignificant in the
subsample of low market-wide valuation.
4.5.2. The RRP effect on the offer premium
The view that investors’ perception is captured by the reference point effect has been examined. It
remains interesting to explore the RRP effect on offer premiums. Baker et al. (2012) show that the
market would look at the firm’s reference point and may perceive that an offer price is a result of
bidder’s overconfidence, given limited information and limited time in which to process the deal. It
is less likely that the reference point effect plays a similar role for the bidders who have more
information about the firm and seems less likely to justify a firm’s valuation according to a single
number.
The first four specifications of Table 4.5 use different categories of control variables.68 The sign and
significance level of the RRP do not change significantly compared with what is reported in the
specification (5), which is the main specification. The RRP is positive and significant at the 1% level
(coefficient 0.089, t = 4.501), showing that a 10% increase in the RRP is associated with an
approximately 0.9% increase in offer premiums. The signs and significance levels of the control
variables are generally consistent with prior M&A studies (Moeller et al., 2004; Alexandridis et al.,
2013). The results obtained by the author suggest that the relatively more overvalued bidders overpay
68 Specification (1) reports the relationship between offer premiums and the reference point, specification (2) controlled
for deal characteristics, specification (3) controlled for deal and bidder characteristics, specification (4) controlled for deal
and target characteristics, specification (5) controlled for all variables. All specifications accounted for year and industry
dummies.
Chapter 4. Relative Reference Prices and M&A Misvaluations
204
for the target according to the RRP, which is consistent with the reference point theory of M&A. The
joint bidder and target reference point effect has an impact on target pricing. It can be interpreted as
follows: bidders who risk being perceived as overvalued have to pay for targets according to the target
reference point to obtain the deal before revising market perceptions, since lower than the target
reference point would decrease the probability of deal success (Baker et al., 2012). Moreover, when
the bidder’s price is close to the 52-week high, managers may find it hard to justify, and will suffer
significant losses in the long run either because of overvaluation to be corrected in the market (Jensen,
2005) or their stocks are to be aggressively sold at around the high market value of the firm (Barberis
and Xiong, 2009). Hence, bidders would pay heavily to revise the market’s misperception. On the
other hand, targets also demand offer premiums based on the bidder reference point, as a lower bidder
reference point might lead targets to believe that bidders are able to afford a higher offer premium.
Given that acquisitions were largely driven by operating synergies and agency problems before 1990,
and by misvaluation after 1990 (Shleifer and Vishny, 2003; Dong et al., 2006), time-distribution tests
whose results are reported in Table 4.6 were carried out. The sample period in this chapter divides
into three based on M&A merger waves. Before 1990, 1990 to 2000 and post 2000. The sample period
covers the high valuation trends when investors’ perception to firms’ misvaluation is high, such as
the stock market bubble between 2000 and 2002, and the housing bubble and credit crisis between
2007 and 2009. It should also be suggested that the misvaluation hypothesis is more likely to explain
acquisitions after 1990 than before 1990 when the primary M&A motive is synergies. The results
yield similar results to our predictions. By extending the sample period of Dong et al. (2006), it was
found that RRP plays an important role during large valuation trends. The sample period divides into
two: from 2001 to 2007 and 2008 to 2014, in order to investigate how RRP explains the offer
Chapter 4. Relative Reference Prices and M&A Misvaluations
205
premiums before and after 2008 financial crisis, as reported in Table 4.7. The results obtained indicate
that for every 10% increase in the RRP an increase of 0.94% in the period of 2001 and 2007 would
occur, this figure more than double during the period of 2008 to 2014. Misvaluation tends to be larger
after financial crisis. Once again, the results the research obtained show that the relative overvaluation
drives M&A overpayment. Unlike Rhodes-Kropf and Viswanathan (2004), who found a positive
correlation between stock-financed acquisitions and merger waves and argue that rational targets
make mistakes by accepting overvalued stocks during high valuation periods, up-to-date evidence is
provided here suggesting that premiums increase with valuation errors. This also rationalises the
bidder’s motive for overpayment, as high offer premiums area form of compensation for the target’s
willingness to accept overvalued stocks.
However, it is important to be aware of the fact that not all acquisitions involve a bidder that is
relatively more overvalued than the target, as reflected in the RRP. The proxy is used to investigate
whether the primary M&A motive is overvaluation from the market perspective. Bidders are
relatively more overvalued if their current price shows smaller changes relative to the reference point
price as opposed to those of the targets (i.e. RRP>0), and they are relatively undervalued if the
changes are larger relative to the reference point price compared with those changes of targets (i.e.
RRP<0). If the RRP is a suitable proxy for misvaluation, it should accommodate the predictions of
the misvaluation hypothesis. Based on this, higher RRP acquisitions associated with higher offer
premiums and target announcement returns, and lower bidder announcement returns than lower RRP
acquisitions, should be expected.
The sample was divided into two subsamples according to the RRP. First, the univariate analysis
Chapter 4. Relative Reference Prices and M&A Misvaluations
206
Results of Table 4.8 are presented. The results show a majority of deals are RRP driven: 1,155 as
opposed to 723. Panel A indicates that acquisitions are carried with significantly larger offer
premiums when bidders are relatively more overvalued than the opposite case, the mean difference
for the offer premium is 5% and at the 1% significance level. Our results suggest overvalued bidders
are likely to pay with high offer premiums for the undervalued or less overvalued targets. Managerial
priority in takeovers is to dilute overvaluation. Because of this, they believe the deal may not be
accepted until high offer premiums paid to targets who can also identify bidders’ motive through RRP.
In Panel B, both the two subsamples show significantly negative bidder announcement returns. The
relatively more overvalued bidders perform significantly worse than the relatively more undervalued
bidders. With respect to target announcement returns, both relatively more undervalued and more
overvalued targets receive high market reactions, which is significantly different from 0. In particular,
the mean difference for target announcement returns is about 5%, suggesting that targets involved in
high RRP acquisitions can demand high offer premiums based on RRP, which is translated into higher
announcement returns.
Panel D of this table reports univariate analysis results for the offer premium by the method of
payment in the two RRP subgroups. The results obtained are consistent with the prediction of the
misvaluation hypothesis that overvalued stocks are used as cheap currency. Specifically, all cash-
financed acquisitions carry higher offer premiums when bidders are relatively more overvalued than
relatively more undervalued bidders, as the result shows a mean difference of 8.5%, which is at 1%
significance level. Moreover, all stock-financed acquisitions carry lower offer premiums than all
cash-financed acquisitions when bidders are relatively more overvalued. Combined, relatively more
overvalued bidders tend to avoid using cash, as it will increase bidders’ takeover costs. Our results
Chapter 4. Relative Reference Prices and M&A Misvaluations
207
indicate that the method of payment indicates clear sign that managers time the market.
Following univariate analyses, multivariate analyses of the offer premium on RRP in two different
RRP subsamples were conducted, as reported in Table 4.9. Specification (1) shows the offer premium
increases 1.64% for every 10% increase of RRP when the bidders are relatively more overvalued.
The results are not significant in the comparable analysis. Our results show bidders to be perceived
as relatively more overvalued are more likely to pay for targets with high offer premiums for the
purpose of revising the market misperception. Our results also show that larger bidders generously
pay for smaller targets, which is consistent with the findings of Moller et al. (2004).
4.5.3. Do bidders who focus on the RRP overpay for the target?
The RRP was used as an instrument variable of the offer premium when examining the role of
overpayment. Baker et al. (2012) suggest the offer premium is not a clean measure since it can
represent both overpayment and synergies. 2SLS enables the investigation of the chain responses of
the RRP effect on the offer premium, and the offer premium effect on firm’s announcement returns.
If the RRP plays a role in overpayment, it should be expected that the offer premium using 2SLS
estimates should yield more negative (positive) bidder (target) announcement returns compared with
those using OLS estimates.
In Table 4.10, the offer premium with 2SLS estimates generates more negative (positive) bidder
(target) CARs compared with that with OLS estimates, shown in specifications (1) and (2), and (3)
Chapter 4. Relative Reference Prices and M&A Misvaluations
208
and (4) respectively.69 The results are consistent with Baker et al. (2012) who studied the reference
point effect on M&A outcomes in a sample of M&As between 1984 and 2007. The results are also in
line with the reference point theory of M&As. The market would presumably believe that the chance
of price rebounds tends to increase when the bidder’s current stock price is close to the 52-week high.
During the time when a bid is announced, the market reacts negatively to it as this may indicate that
bidders are unable to deliver real support to the firm’s performance and they are likely to undertake
bad acquisitions to maintain the overvaluation (Jensen, 2005).
4.5.4. Do all stock-financed acquisitions driven by the RRP protect the wealth of long-term
shareholders?
Thus far, the proposition that relatively more overvalued bidders are more likely to use stocks as a
means of payment and tend to pay significantly higher offer premiums using the new misvaluation
measure of the RRP has been examined. Now the issue of whether or not bidders focusing on RRP
protect long-term shareholders’ interest will be investigated. According to the misvaluation
hypothesis, bidders who dilute overvaluation with stocks attempt to protect the wealth of long-term
shareholders. Thus, it should be reasonable to expect that bidders making an offer price based on the
RRP. In Panel A of Table 4.11, the sample was limited to acquisitions that are 100% financed by
stocks only and the sample was ranked into four quartiles according to the RRP, each presenting 152
observations.70 The aim was to examine whether overpayment leads to underperformance. By doing
so, the offer premiums and long-term performance under the market-adjusted model for each
correspondent quartile were estimated. The fourth quartile (i.e. the highest quartile) includes
69 The Hausman test and F-test results show that coefficients generated by 2SLS regression are more consistent with those
generated by OLS regression, as reported at the end of Table 4.10.
70 Of the 608 all stock-financed acquisitions, 402 fall into the group in which bidders are relatively more overvalued.
Chapter 4. Relative Reference Prices and M&A Misvaluations
209
acquisitions involving relatively more overvalued bidders whereas the other quartiles include
acquisitions involving relatively less overvalued or more undervalued bidders.
It was also found that stock-financed acquisitions generate negative long-term returns, which are
consistent with the M&A literature (Loughran and Vijh, 1997; Rau and Vermaelen, 1998). Bidders in
the highest quartile pay the highest offer premiums compared with those of other quartiles. In spite
of this, the mean difference of offer premiums between the relative more overvalued bidders and
undervalued bidders is 9.9% at 1% significance level and the mean difference for long-term abnormal
returns of stock bidders is 18.7% at 10% significance level, indicating that relatively more overvalued
bidders outperform their undervalued counterparts. Combined, this researcher’s results suggest that
bidders paying high offer premiums according to the RRP are able to protect the wealth of long-term
shareholders. These results are consistent with Ang and Cheng (2006) who found that the long-term
performance of stock bidders and overvaluation are positively related. However, the researcher’s
findings contradict to those of Lin et al. (2011). Their paper classified bidder valuation by the ratio
of price-to-fundamental value (P/V) with higher P/V indicating a more overvalued bidder, and found
that bidders who have the highest P/V generate a significant negative market performance for short
term and long term within three years following M&As as compared with those bidders in the other
P/V quartiles.
For the purpose of robustness checks, the long-term performance of stock bidders with the size-
adjusted model was also examined, and stock bidders were divided into three ranks according to the
RRP, as shown in Panel B of the table. This researcher’s results show that the relatively more
overvalued bidders tend to pay 36% offer premiums for targets, which is significantly 6.6% higher
Chapter 4. Relative Reference Prices and M&A Misvaluations
210
than those paid by relatively more undervalued bidders. It is also evident that the long-term
performance of those paying significantly higher offer premiums is better. The results are quite similar
to those presented in Panel A of the table.
4.6. Robustness checks
4.6.1. Endogeneity issues
OLS can be subject to endogeneity issues arising from omitted variable biases in this chapter, as the
RRP maybe correlated with firms’ mismanagement or mispricing which cannot be observed or the
possibility that the market perception is likely the be an accurate reflection of the firm’s valuation. If
the market could accurately estimate the value of the firm, the managers should have no chance to
time the market by means of the RRP. However, this tends to be unrealistic as the misvaluation
hypothesis proposes. In this case, in this paper it is suggested that the market’s 52-week high and the
bidder’s and target’s 65-week high can be used as instrumental variables given that they are indirectly
related to offer premiums but can affect offer premiums through the RRP. The market’s 52-week high
is an ideal instrumental variable as it is uncorrelated with a firm’s mismanagement. It is suggested
that the bidder’s and the target’s 65-week highs are used as instrumental variables in line with the
extrapolation hypothesis that indicates that market perception is influenced by firms’ past
performance. Using a longer horizon is of less relevance in terms of the market perception of a firm’s
valuation. A 2SLS estimator was therefore performed, with the market 52-week high and the bidder
and the target 65-week high as instrumental variables.
Chapter 4. Relative Reference Prices and M&A Misvaluations
211
According to the results obtained, shown in Table 4.12, the OLS is preferred over the 2SLS.71 This is
because the market 52-week high reflects the market-wide valuation instead of the firm-specific
valuation which is believed to be an important source of valuation error (Rhodes-Kropf et al., 2005).
The results indicate that the OLS is likely to dominate the 2SLS. In addition to this, a correlation
matrix was performed, as reported in Table 4.13. The obtained results show little evidence of
econometric problems, such as multicollinearity issues.
4.6.2. The effect of the RRP on the probability of using stocks
Results of a series of robustness checks examining the RRP effect on the probability of using stocks
are reported in Table 4.14. Shleifer and Vishny (2003) predicted that bidders use stocks when they
are relatively more overvalued. The M&A sample in this chapter was divided into two sub-groups:
the relatively more overvalued bidders (i.e. RRP>0), and the relatively more undervalued bidders (i.e.
RRP<0).72 The results obtained suggest that the RRP effect on the probability of using stocks is solely
driven by the relatively more overvalued bidders. It was found that, when the bidders are relatively
more overvalued, every one unit increases in the RRP increases the propensity for using stocks for
acquisitions by 16.4 percentage points, whereas there is little evidence that the relatively more
undervalued bidders use stocks for acquisitions. Table 4.15 reports that the probability of using stocks
rather than other means of payment is also large when the RRP increases. As seen, stocks are more
likely to be used as a means of payment for acquisitions compared with the method of payment as
71 Specifically, the Hausman test shows a p-value of 0.6414, indicating there are no endogeneity issues in the regression.
Moreover, the test of over-identifying restrictions further rejects the null hypothesis that there is no relation between the
instruments and the error term (p = 0.0000).
72 1,742 observations in regressions were recorded. Of these observations, 1,103 fall into the group in which bidders are
relatively more overvalued and 639 fall into the group in which bidders are relatively more undervalued. 50 observations
were dropped due to multicollinearity problems in the industries.
Chapter 4. Relative Reference Prices and M&A Misvaluations
212
cash and a mixture of stocks and cash respectively. 73 The results of additional controls are consistent
with the main results of this chapter presented in Table 4.3.
4.6.3. The effect of the RRP on the offer premium
This section reports the results of subsample robustness tests relating to the RRP effect on the offer
premium. The sample was divided into subsamples according to the method of payment, deal type,
deal choice and deal status in Table 4.16 to 4.19 respectively, and control variables are the same as
those in Table 4.5. Table 4.16 reports a positive relationship between the RRP and the offer premiums
by different methods of payment subsamples. The offer premium is larger when paying with cash
than stocks, which is consistent with the univariate analysis results presented in Panel D of Table 4.8.
It is worth noting that in stock-financed subsample, bidders tend to pay generously for those with a
higher ROA, implying that bidders tend to select firms with strong profit-generating ability, likely to
create value for the combined firm. The results obtained show that bidders are likely to pay with large
offer premiums regardless of the method of payment choices. Table 4.17 reports that an increase in
the distance of the RRP increases the offer premium regardless of deal types. Signs and significance
levels for additional controls are quite similar between the two subsamples. In Table 4.18, the offer
premium for merger subsamples is larger than for tender offers. It can be argued that a negotiation
process between the two firms involved may incorporate more private information about the firms,
leading the RRP to be well interpreted. In that case, bidders can more easily justify their M&A motives
as well as the target value with target managers rather than shareholders, thus resulting in lower offer
premiums. The sample used also includes a small proportion of unsuccessful deals as opposed to
73 Our results continue to hold by replacing the binary variable with a continuous variable. This allows us to observe how
the RRP affects the proportion of stocks used in the payment structure. We omitted this table for the sake of brevity.
Chapter 4. Relative Reference Prices and M&A Misvaluations
213
successful deals. In Table 4.19, the RRP effect is strong for both subsamples.74 On the whole, the
results indicate that the RRP have a strong effect on offer premiums.
4.6.4. Do bidders who focus on RRP overpay for the target? Testing with the market-adjusted model
So as to avoid measurement error due to the model employed for the firms’ announcement returns in
analysing the role of the RRP played M&As, the market-adjusted model was used to calculate the
firms’ announcement returns, as reported in Table 4.20. These results are consistent with those in the
market model. Specification (1) reported that bidders’ CARs decrease by 1.4% for a one percent
increase in the offer premium, which is slightly higher than those predicted by the market model (-
1.8%). On the other hand, specification (3) indicated that targets’ CARs increase by 46.9% for every
one percent increase in the offer premium, which is as the same as that predicted by the market model.
A 2SLS estimator was also conducted using the RRP as an instrumental variable of the offer premium
in the regression of CARs on the offer premium, with a similar reason has been explained in the main
regression analysis. Including the RRP effect in the 2SLS, the results show a more positive bidder
CARs and more negative target CARs as compared with those under OLS. Once again, paying
according to the RRP impresses the investors’ minds that the deal is overpaid.
4.7. Conclusion
This chapter has investigated the misvaluation hypothesis using the reference point theory of M&As.
A bilateral valuation framework with RRP has been built, and results are consistent with what the
misvaluation hypothesis of Shleifer and Vishny (2003) predicts.
74 This was used as a robustness test rather than to interpret the RRP effect as the reason for deal completion, as what
determines a deal completion is complex.
Chapter 4. Relative Reference Prices and M&A Misvaluations
214
The target and the bidder reference points have been added into the M&A platform and have found
that the propensity for paying for acquisitions with stocks is greater when the RRP increases, this
trend being more pronounced when the market misperception of a firm’s valuation is high. The results
obtained are consistent with those of Rhodes-Kropf et al. (2005). Moreover, the offer premium
increases with RRP, leading to more negative bidder (positive target) announcement returns. The
results indicate that RRP plays an overpayment role. In a quartile analysis for a sample of all stock-
financed acquisitions, it can be found that the relatively more overvalued bidders who pay highest
offer premiums compared with those in other quartiles generate less negative long-term abnormal
returns, suggesting that bidders time the market while focusing on the RRP.
Several contributions have been made to behavioural finance and M&A literature. First, a simply way
of structuring M&A through the RRP was built, which reflects the most current market reactions to a
firm’s valuations. This valuation measure is straightforward in terms of observing the difference
between bidder and target overvaluation. The market tends to react to bidders’ announcements
negatively, as there is a high offer premium paid according to the RRP. Secondly, it has been found
that the RRP directs the method of payment choices, as a relatively more overvalued bidder tends to
reduce offer premiums when financing an acquisition with stocks rather than with cash. Thirdly, it
has also been revealed that bidders who have more information about the firm are also subject to the
reference point effect. However, this does not mean that bidders are irrational, rather, the findings
show that higher offer premiums according to the RRP would mitigate negative market reactions in
the long term, suggesting that focusing on the RRP is a bidder’s way of thinking weighed towards the
interests of the long-term shareholders. Overall, the results of this chapter suggest that managers are
able to time the market through the RRP, which is consistent with the misvaluation hypothesis.
Chapter 4. Tables
215
Table 4.1: Summary statistics for M&A sample
This table reports summary statistics for 1,878 U.S. domestic public deals announced between 1985 and 2014. The number
N denotes the number of deals per year. The third and fourth columns present the mean and median of deal value. The
fifth to seventh columns present the method of payment. Here “Stock” (“Cash”) refers to all-stock (all-cash) acquisitions.
“Mix” refers to acquisitions that are neither all stock-financed nor all cash-financed acquisitions. “Completed” refers to
completed deals (i.e. successful deals), of which there are 1,874 with information relating to deal status. “Tender” refers
to tender offers. “Hostile” refers to hostile bids. “Diversified” refers to diversified deals in which the primary two Standard
Industry Classification codes are different between bidders and targets.
Year N Deal Value ($mils) Payment Method Completed Tender Hostile Diversified
Mean Median Cash Stock Mix Y N Y N Y N Y N
1985 5 243.86 53.00 3 1 1 3 2 - 5 1 4 2 3
1986 21 129.04 41.70 18 1 2 18 3 7 14 1 20 15 6
1987 34 254.60 47.25 18 6 8 30 3 6 28 5 29 17 17
1988 30 381.53 68.92 15 8 7 24 6 11 19 6 24 16 14
1989 24 114.04 30.49 11 10 3 17 6 4 20 1 23 18 6
1990 23 579.03 29.38 12 7 4 18 5 5 18 3 20 10 13
1991 23 172.38 26.82 7 14 1 19 3 3 20 - 23 14 9
1992 21 155.46 51.44 6 13 2 17 4 1 20 2 19 12 9
1993 45 519.07 114.00 14 18 12 33 12 7 38 5 40 20 25
1994 61 222.95 74.12 23 30 8 46 14 9 52 7 54 18 43
1995 98 538.98 74.91 27 53 18 84 14 14 84 6 92 38 60
1996 95 684.53 138.25 31 40 24 80 15 18 77 9 86 36 59
1997 130 645.10 232.11 19 62 49 113 17 19 111 3 127 46 84
1998 138 1208.89 140.12 36 54 47 126 12 27 111 3 135 47 91
1999 164 1513.65 305.42 61 58 45 134 30 41 123 14 150 68 96
2000 136 2286.88 378.34 35 68 33 119 17 32 104 8 128 49 87
2001 109 1115.01 146.89 33 40 36 94 15 27 82 4 105 38 71
2002 49 1784.72 268.90 20 14 15 44 5 16 33 4 45 18 31
2003 71 807.01 130.82 27 18 26 65 6 19 52 5 66 19 52
2004 66 2859.54 479.02 25 16 25 60 6 6 60 3 63 22 44
2005 66 2874.25 500.75 29 13 24 60 6 7 59 5 61 24 42
2006 75 1838.00 563.07 40 12 23 65 10 6 69 4 71 29 46
2007 60 1478.29 792.51 39 6 15 53 7 12 48 2 58 18 42
2008 57 2208.95 234.26 35 6 16 40 17 14 43 10 47 17 40
2009 44 3498.35 496.88 17 8 19 41 3 16 28 - 44 18 26
2010 56 1884.97 572.72 33 9 14 47 9 16 40 5 51 15 41
2011 41 2691.37 611.62 16 8 17 28 13 8 33 9 32 13 28
2012 38 1385.71 622.51 26 1 11 37 1 10 28 - 38 17 21
2013 44 1997.05 1139.09 28 5 11 39 5 9 35 3 41 12 32
2014 54 6908.66 1662.39 22 9 23 43 11 10 44 6 48 16 38
Total 1878 1540.29 227.49 726 608 539 1597 277 380 1498 134 1744 702 1176
Chapter 4. Tables
216
Table 4.2: Summary statistics for variables
This table reports the number, mean, median and standard deviation of variables used in the regressions. Panel A presents the main
dependent variables. Firms’ 3-day CARs are calculated with the market model, with parameters estimated between 261 and 28 trading
days prior to the announcement date. Offer premiums are defined as the logarithmic term difference between offer price and target
stock price 30 calendar days prior to the announcement date. Panel B presents the main variables of interest. The reference point is
defined as the logarithmic term difference between a firm’s highest stock price over 335 calendar days ending 30 days prior to the
announcement date and the price ending 30 days prior to the announcement date. RRP is defined as the difference between the target
and the bidder reference point. Panel C presents control variables. Deal characteristics are noted as Table 1. “Relative Size” is defined
as the deal value divided by bidder MV, where bidder MV is defined as the product of market price and outstanding shares (CRSP:
SHROUT*PRC). ROA is return-on-asset ratio, defined as net income (Compustat: NI) divided by total asset (Compustat:AT). MTBV
is market-to-book value, defined as the market value of equity to the book value of equity, where book value of equity is total
shareholders’ equity (Compustat: SEQ) plus deferred taxes and investment tax credit (Compustat: TXDITC) minus the preferred stock
redemption value (Compustat: PSTKRV). Volatility is the standard deviation of daily returns over the 335 calendar days ending 30
days prior to the announcement date. CF/E is cash flow-to-equity ratio, defined as income before extraordinary items (Compustat: IBC)
plus depreciation and amortization (Compustat: DPC) minus cash dividends (Compustat: DV), and leverage is measured by debt-to-
equity ratio, defined as total long-term debt (Compustat: DITT) divided by the book value of equity. All accounting variables were in
the fiscal year end before the announcement date, and continuous variables are winsorised at the 1% and 99% level.
Variables N Mean Median Std. Dev.
Panel A: Main Dependent Variables
Offer Premiums 1878 0.310 0.292 0.282
Bidder 3-day CARs 1878 -0.011 -0.007 0.073
Target 3-day CARs 1878 0.222 0.176 0.240
Panel B: Main Variables of Interest
Bidder Reference Point 1878 0.294 0.157 0.351
Target Reference Point 1878 0.412 0.255 0.455
RRP 1878 0.118 0.055 0.420
Panel C: Other Variables: deal, bidder, and target characteristics
Cash 1878 0.387 - 0.487
Stock 1878 0.324 - 0.468
Hostile 1878 0.071 - 0.257
Tender 1878 0.202 - 0.402
Diversification 1878 0.374 - 0.484
Relative Size 1878 0.401 0.191 0.555
Bidder lnMV 1878 7.500 7.443 2.159
Bidder MTBV 1878 3.991 2.632 6.269
Bidder ROA 1878 0.026 0.047 0.138
Bidder Volatility 1878 0.029 0.025 0.017
Bidder Leverage 1872 0.661 0.330 1.194
Bidder CF/E 1804 0.161 0.180 0.316
Target lnMV 1878 5.212 5.133 1.839
Target MTBV 1878 2.691 1.797 4.300
Target ROA 1878 -0.049 0.024 0.241
Target Volatility 1878 0.041 0.035 0.022
Target Leverage 1870 0.619 0.140 1.621
Chapter 4. Tables
217
Table 4.3: The effect of the RRP on the probability of using stocks
This table reports binomial logistic regression for all-stock acquisitions on RRP. Dependent variable is “Stock”, which is a dummy variable, taking a value of 1 if acquisitions are 100%
financed with stocks, 0 otherwise. Variable definitions are as in the notes to Table 4.2. P-value is reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is
denoted ***, ** and * respectively, is reported alongside marginal effects. We transfer coefficients into marginal effect (ME), evaluated at the sample means of the independent variables.
(1) (2) (3) (4)
Stock Coef. p-value ME Coef. p-value ME Coef. p-value ME Coef. p-value ME
RRP 0.567*** (0.000) 0.103 0.605*** (0.000) 0.110 0.531*** (0.001) 0.097 0.570*** (0.001) 0.104
Hostile -0.777** (0.014) -0.141 -0.742** (0.019) -0.135 -0.721** (0.022) -0.132 -0.670** (0.033) -0.123
Tender -2.441*** (0.000) -0.443 -2.402*** (0.000) -0.436 -2.563*** (0.000) -0.468 -2.502*** (0.000) -0.458
Diversification 0.146 (0.266) 0.027 0.124 (0.349) 0.023 0.146 (0.280) 0.027 0.120 (0.378) 0.022
RelativeSize -0.611*** (0.000) -0.111 -0.662*** (0.000) -0.120 -0.561*** (0.000) -0.102 -0.606*** (0.000) -0.111
Bidder lnMV -0.379*** (0.000) -0.069 -0.290*** (0.000) -0.053 -0.402*** (0.000) -0.073 -0.314*** (0.000) -0.057
Bidder MTBV 0.048*** (0.000) 0.009 0.036*** (0.001) 0.007 0.060*** (0.000) 0.011 0.048*** (0.000) 0.009
Bidder ROA -2.063*** (0.000) -0.375 -0.914* (0.088) -0.166 -2.464*** (0.000) -0.450 -1.407** (0.030) -0.257
Target lnMV 0.315*** (0.000) 0.057 0.350*** (0.000) 0.063 0.348*** (0.000) 0.064 0.375*** (0.000) 0.069
Target MTBV 0.051*** (0.001) 0.009 0.041*** (0.006) 0.008 0.063*** (0.000) 0.012 0.055*** (0.001) 0.010
Target ROA 0.087 (0.762) 0.016 0.371 (0.231) 0.067 0.009 (0.977) 0.002 0.218 (0.492) 0.040
Target Volatility 5.118 (0.253) 0.928 3.773 (0.423) 0.690
Bidder Volatility 32.029*** (0.000) 5.809 30.994*** (0.000) 5.670
Bidder Leverage -0.232*** (0.001) -0.042 -0.211*** (0.002) -0.039
Target Leverage -0.126*** (0.003) -0.023 -0.121*** (0.005) -0.022
Bidder CF/E 0.213 (0.403) 0.039 0.233 (0.386) 0.043
Year Yes Yes Yes Yes
Industry Yes Yes Yes Yes
Constant 0.324 (0.577) -1.480** (0.032) 0.592 (0.360) -1.130 (0.132)
N 1878 1878 1792 1792
Pseudo R2 0.272 0.287 0.291 0.303
Chapter 4. Tables
218
Table 4.4: The effect of the RRP on the probability of using stocks under different market-wide valuations
This table reports binomial logistic regression for all-stock acquisitions on RRP by different market conditions. Dependent variable is “Stock”, which is a binary variable, taking value
of 1 if acquisitions are financed with 100% stocks, 0 otherwise. High Market is a dummy variable, taking a value of 1 if takeover months in the top 25% above past 5-year average de-
trended P/E of the market index (S&P 500) or market valuation is high, 0 otherwise. Specifications (1)-(3) report the results when market-wide valuation is high, low and neutral
respectively. Due to the availability of the P/E of the market index, we were able to determine 1733 observations with valid market-wide valuation data, 666 for high valuation periods,
434 for low valuation periods, and 633 for neutral valuation periods. Variable definitions are as in the notes to Table 4.2. P-value is reported in parentheses. Statistical significance at
the 1%, 5% and 10% levels is denoted ***, ** and * respectively, is reported alongside marginal effects. We transfer coefficients into marginal effect (ME), evaluated at the sample
means of the independent variables.
Chapter 4. Tables
219
Table 4.4. (continued)
(1) (2) (3)
Stock High Low Neutral
Coef. p-value ME Coef. p-value ME Coef. p-value ME
RRP 0.758*** (0.002) 0.153 1.021 (0.113) 0.059
0.510* (0.057) 0.109
Hostile -0.317 (0.462) -0.064 -1.043 (0.199) -0.060
-0.723 (0.332) -0.155
Tender -2.339*** (0.000) -0.473 -1.195** (0.042) -0.069
-4.251*** (0.000) -0.910
Diversification 0.797*** (0.001) 0.161 -0.165 (0.676) -0.010
-0.060 (0.792) -0.013
RelativeSize -0.968*** (0.000) -0.196 -0.847 (0.121) -0.049
-0.307 (0.238) -0.066
Bidder lnMV -0.377*** (0.000) -0.076 -0.809*** (0.000) -0.047
-0.141 (0.162) -0.030
Bidder MTBV 0.052*** (0.002) 0.011 0.060 (0.181) 0.003
0.038 (0.109) 0.008
Bidder ROA -2.376*** (0.009) -0.481 -1.664 (0.453) -0.096
-2.797* (0.071) -0.599
Target lnMV 0.600*** (0.000) 0.121 0.696** (0.014) 0.040
0.170 (0.161) 0.036
Target MTBV 0.062** (0.016) 0.013 -0.004 (0.936) 0.000
0.063** (0.049) 0.014
Target ROA -0.154 (0.764) -0.031 0.069 (0.938) 0.004
0.699 (0.219) 0.150
Target Volatility 8.929 (0.254) 1.808 -1.335 (0.952) -0.077
-1.354 (0.866) -0.290
Bidder Volatility 15.449 (0.144) 3.127 35.750 (0.140) 2.068
51.081*** (0.000) 10.936
Bidder Leverage -0.306*** (0.002) -0.062 -0.219 (0.304) -0.013
-0.169 (0.150) -0.036
Target Leverage -0.129* (0.090) -0.026 -0.089 (0.350) -0.005
-0.203** (0.013) -0.043
Bidder CF/E 0.096 (0.791) 0.019 -0.634 (0.359) -0.037
2.044*** (0.001) 0.438
Year Yes Yes Yes
Industry Yes Yes Yes
Constant -2.916** (0.033) -0.057 (0.974) 0.044 (0.976)
N 666 434 633
Pseudo R2 0.345 0.268 0.333
Chapter 4. Tables
220
Table 4.5: The effect of the RRP on offer premiums
This table reports the ordinary least square (OLS) regression results for offer premiums on the RRP, controlling for a series of deal and firm characteristics. Specification (1) reports
the relationship between offer premiums and the reference point effect, specification (2) controls for deal characteristics, specification (3) controls for deal and bidder characteristics,
specification (4) controls for deal and target characteristics, specification (5) controls for all variables. Variable definitions are as in the notes to Table 4.2. Robustness t-statistics are
reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is denoted ***, **and * respectively, is reported alongside coefficients.
Offer Premiums (1) (2) (3) (4) (5) Coef. t-stat. Coef. t-stat. Coef. t-stat. Coef. t-stat. Coef. t-stat.
RRP 0.101*** (5.099) 0.105*** (5.311) 0.110*** (5.575) 0.098*** (4.846) 0.089*** (4.501)
Hostile 0.008 (0.367) 0.002 (0.105) 0.015 (0.698) 0.032 (1.563)
Tender 0.075*** (4.744) 0.074*** (4.695) 0.080*** (5.098) 0.075*** (5.006)
Diversification 0.006 (0.433) 0.004 (0.284) 0.005 (0.369) -0.017 (-1.256)
Stock -0.024 (-1.336) -0.019 (-0.991) -0.026 (-1.392) -0.003 (-0.184)
Cash -0.009 (-0.565) -0.004 (-0.227) -0.019 (-1.123) -0.031* (-1.917)
RelativeSize 0.020 (1.537) 0.021* (1.744) 0.116*** (7.263)
Bidder ROA 0.090 (1.262) 0.019 (0.269)
Bidder MTBV -0.000 (-0.222) -0.000 (-0.010)
Bidder lnMV 0.006 (1.506) 0.056*** (9.627)
Bidder Volatility 0.570 (0.819) 0.768 (1.047)
Target ROA 0.119*** (2.810) 0.141*** (3.347)
Target MTBV -0.001 (-0.806) -0.001 (-0.878)
Target lnMV -0.030*** (-6.512) -0.078*** (-11.703)
Target Volatility -0.035 (-0.065) -0.315 (-0.537)
Constant 0.212*** (3.291) 0.182*** (2.744) 0.116 (1.535) 0.334*** (4.388) 0.150* (1.896)
Year Yes Yes Yes Yes Yes
Industry Yes Yes Yes Yes Yes
N 1878 1878 1878 1878 1878
adj. R2 0.077 0.088 0.089 0.116 0.169
Chapter 4. Tables
221
Table 4.6: The effect of the RRP on offer premiums over time
This table reports the OLS regression results for offer premiums on the RRP by time periods. We divided our sample into
different time periods. Specification (1)-(3) report results before 1990, 1990 to 2000, and after 2000 respectively. Variable
definitions are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses. Statistical significance at
the 1%, 5% and 10% levels is denoted ***, **and * respectively, is reported alongside coefficients. (1) (2) (3)
Offer Premiums 1985-1989 1990-2000 2001-2014
Coef. t-stat. Coef. t-stat. Coef. t-stat.
RRP -0.058 (-0.773) 0.062** (2.392) 0.125*** (3.858)
Hostile -0.049 (-0.463) 0.044 (1.476) 0.031 (1.038)
Tender 0.191*** (2.750) 0.109*** (4.668) 0.048** (2.152)
Diversification -0.021 (-0.395) -0.003 (-0.149) -0.024 (-1.255)
Stock 0.078 (0.931) -0.002 (-0.076) -0.039 (-1.353)
Cash 0.037 (0.564) -0.084*** (-3.258) 0.006 (0.268)
RelativeSize 0.064 (1.413) 0.133*** (5.774) 0.092*** (3.355)
Bidder ROA 0.623** (2.306) -0.010 (-0.104) -0.005 (-0.044)
Bidder MTBV 0.001 (0.273) -0.001 (-0.432) 0.001 (0.848)
Bidder lnMV 0.028 (1.544) 0.058*** (6.615) 0.049*** (5.472)
Bidder Volatility -4.431 (-1.110) 0.534 (0.535) 0.733 (0.611)
Target ROA 0.157 (0.911) 0.188*** (2.996) 0.121** (1.979)
Target MTBV -0.004 (-0.378) -0.000 (-0.040) -0.003 (-1.577)
Target lnMV -0.032 (-1.508) -0.079*** (-7.832) -0.074*** (-7.090)
Target Volatility 1.856 (0.774) -0.330 (-0.415) 0.093 (0.098)
Constant -0.413** (-2.600) 0.087 (0.860) 0.301*** (3.104)
Year Yes Yes Yes
Industry Yes Yes Yes
N 114 934 830
adj. R2 0.190 0.176 0.184
Chapter 4. Tables
222
Table 4.7: The effect of the RRP on offer premiums by valuation periods
This table reports the OLS regression results for offer premiums on the RRP before and after the 2008 financial crisis. The sample was divided into two subsamples: between 2000 and
2007 and between 2008 and 2014, i.e. financial crisis period, presented in specification (4) and (5) respectively. Variable definitions are as in the notes to Table 4.2. Robustness t-
statistics are reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is denoted ***, **and * respectively, is reported alongside coefficients.
(1) (2)
Offer Premiums 2001-2007 2008-2014
Coef. t-stat. Coef. t-stat.
RRP 0.094** (2.293) 0.192*** (3.655)
Hostile 0.089** (2.245) -0.003 (-0.073)
Tender 0.017 (0.539) 0.080** (2.514)
Diversification -0.046* (-1.772) 0.010 (0.306)
Stock -0.005 (-0.142) -0.081* (-1.854)
Cash 0.023 (0.784) 0.004 (0.138)
RelativeSize 0.069* (1.892) 0.121*** (2.981)
Bidder ROA -0.083 (-0.694) 0.300 (1.273)
Bidder MTBV 0.001 (0.462) 0.003 (1.059)
Bidder lnMV 0.037*** (3.633) 0.059*** (3.631)
Bidder Volatility 0.424 (0.279) 0.799 (0.375)
Target ROA 0.151** (2.176) 0.011 (0.097)
Target MTBV 0.001 (0.238) -0.006** (-2.322)
Target lnMV -0.067*** (-5.149) -0.076*** (-4.427)
Target Volatility -0.192 (-0.140) -0.026 (-0.024)
Constant 0.285** (2.369) 0.102 (0.685)
Year Yes Yes Industry Yes Yes N 496 334 adj. R2 0.190 0.375
Chapter 4. Tables
223
Table 4.8: Univariate analysis by different RRP groups
This table reports univariate analysis results for the offer premium, the bidder and the target’s CAR3 calculated with market model. We divide our sample into those RRP less than 0,
which is the bidder is relatively more undervalued than the target and those RRP larger than 0, which is the bidder is relatively more overvalued than the target. Panel A reports the
univariate analysis results for the offer premium. Panel B reports the univariate analysis results for the bidder CAR3. Panel C reports the univariate analysis results for the target CAR3.
Panel D reports univariate analysis for the offer premium by the method of payment. Here “Cash” represents that acquisitions are 100% financed with cash. “Stock” represents
acquisitions that are 100% financed by stocks. Specifications (3) and (4) report offer premiums of 100% cash-financed acquisitions at RRP>0 and RRP<0 groups. Specifications (3)
and (4) report offer premiums of 100% stock-financed acquisitions at RRP>0 and RRP<0 groups. The mean value, t-statistics, and the number of observations for the offer premium,
the bidder and the target 3-day abnormal returns around the announcement date are reported in each Panel. The mean difference of t-tests is reported at the end of each Panel. T-statistics
are reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is denoted ***, ** and * respectively.
Panel A: Univariate analysis for the offer premium
Offer Premiums t-stat. N
(1) RRP<0 0.279*** (30.20) 723
(2) RRP>0 0.329*** (37.26) 1,155
Mean difference (2)-(1) 0.050*** (-3.79)
Panel B: Univariate analysis for the bidder CAR3
Bidder CAR3mm t-stat. N
(1) RRP<0 -0.006** (-2.27) 723
(2) RRP>0 -0.014*** (-6.37) 1,155
Mean difference (2)-(1) -0.008** (-.2.28)
Panel C: Univariate analysis for the target CAR3
Target CAR3mm t-stat. N
(1) RRP<0 0.191*** (23.81) 723
(2) RRP>0 0.240*** (32.34) 1,155
Mean difference (2)-(1) 0.049*** (4.33)
Panel D: Univariate analysis for the offer premium by the method of payment at different RRP subgroups
(1) (2) (3) (4) Mean difference RRP>0 RRP<0 RRP>0 RRP<0 (1)-(2) (3)-(4) (1)-(3) (2)-(4) Cash Stock
Offer Premiums (%) 35.25*** 26.75*** 31.33*** 29.22*** 8.50*** 2.11 3.92* -2.47
t-stat (25.18) (20.90) (19.07) (14.51) (4.24) (0.80) (1.88) (-1.08)
N 435 291 402 206 724 608 837 497
Chapter 4. Tables
224
Table 4.9: OLS regressions of offer premiums on the RRP: By RRP subsamples
This table reports OLS regression results for the offer premium on RRP. We divide the sample according to the RRP.
Column (1) presents the results for the offer premium on RRP when the bidder is relatively more overvalued than the
target, column (2) presents the results for the offer premium on RRP when the bidder is relatively undervalued than the
target. Variable definitions are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses. Statistical
significance at the 1%, 5% and 10% levels is denoted ***, **and * respectively, is reported alongside coefficients.
(1) (2)
Offer Premiums RRP>0 RRP<0
Coef. t-stat. Coef. t-stat.
RRP 0.164*** (4.766) -0.005 (-0.112)
Hostile 0.003 (0.105) 0.063** (1.968)
Tender 0.090*** (4.317) 0.055*** (2.724)
Diversification 0.009 (0.483) -0.058*** (-2.923)
Stock -0.004 (-0.170) 0.004 (0.159)
Cash -0.024 (-1.080) -0.036 (-1.548)
RelativeSize 0.128*** (5.132) 0.104*** (5.109)
Bidder ROA -0.050 (-0.577) 0.103 (0.881)
Bidder MTBV 0.002 (1.258) -0.002 (-1.129)
Bidder lnMV 0.051*** (6.456) 0.059*** (6.943)
Bidder Volatility -0.763 (-0.750) 2.070* (1.875)
Target ROA 0.168*** (3.439) 0.115 (1.467)
Target MTBV -0.002 (-0.900) 0.000 (0.119)
Target lnMV -0.079*** (-8.750) -0.073*** (-7.066)
Target Volatility -0.390 (-0.535) -0.756 (-0.736)
Constant 0.060 (0.574) 0.012 (0.078)
Year Yes Yes Industry Yes Yes
N 1155 723
R2 0.220 0.214
Chapter 4. Tables
225
Table 4.10: Do bidder focusing on the RRP overpay for the target?
This table reports the OLS regression and 2SLS regression results for both bidder and target 3-day CARs on offer
premiums. The dependent variable for specifications (1) and (2) is the bidder CAR3 and for specifications (3) and (4) is
the target CAR3, where CAR3 is calculated with the market model. The parameters are estimated in the window [-261,-
28]. Specifications (2) and (4) reports the results of 2SLS regression using the RRP as an instrument variable of offer
premiums. Variable definitions are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses.
Statistical significance at the 1%, 5% and 10% levels is denoted ***, **and * respectively. The results of the Hausman
test and F-test were reported at the lower part of the table.
(1) (2) (3) (4) Bidder CAR3 Target CAR3
OLS IV OLS IV
Offer Premiums -0.018*** -0.090*** 0.469*** 0.765*** (-2.807) (-2.626) (20.499) (6.597)
Hostile -0.002 -0.002 -0.000 -0.003 (-0.324) (-0.317) (-0.028) (-0.153)
Tender 0.009** 0.017*** 0.041*** 0.010 (2.376) (2.888) (3.102) (0.533)
Diversification 0.003 0.003 0.004 0.004 (0.805) (0.867) (0.375) (0.373)
Stock -0.011** -0.010** -0.045*** -0.045*** (-2.326) (-2.155) (-3.952) (-3.512)
Cash 0.021*** 0.021*** 0.031*** 0.037*** (5.245) (4.555) (2.615) (2.857)
RelativeSize -0.005 -0.005 -0.050*** -0.052*** (-1.116) (-1.311) (-6.136) (-5.505)
Bidder lnMV -0.003*** -0.003*** (-3.916) (-3.230)
Bidder MTBV -0.000 -0.000 (-0.972) (-0.901)
Bidder ROA 0.041** 0.040*** (2.186) (2.998)
Target lnMV -0.002 0.006 (-0.866) (1.447)
Target MTBV -0.002** -0.002 (-2.228) (-1.640)
Target ROA -0.012 -0.017 (-0.469) (-0.800)
Constant 0.015* 0.032*** 0.106*** -0.027 (1.843) (2.706) (5.578) (-0.483)
N 1878 1878 1878 1878
adj. R2 0.060 . 0.374 .
Hausman test X2=4.8934 (p=0.0271) X2=7.9056 (p=0.0050)
F-test X2=60.2985 (p=0.0000) X2=42.9958 (p=0.0000)
Chapter 4. Tables
226
Table 4.11: Do all stock-financed acquisitions driven by the RRP protect the wealth of long-term shareholders?
Panel A of this table reports the univariate analysis results of both the offer premium and the firms’ 36-month market-
adjusted buy-and-hold abnormal returns (BHAR36ma) by RRP quartiles. The sample only consists of 100% stock-
financed acquisitions. Each quartile is assigned a rank from 1 to 4. Rank 1 represents bidders that are relatively more
undervalued than their targets (i.e. RRP<0), and rank 4 represents bidders that are relatively more overvalued than their
targets (i.e. RRP>0). Panel B of this table reports the univariate analysis results of both the offer premium and the firms’
36-month size-adjusted buy-and-hold abnormal returns (BHAR36sa) by the RRP. We divided the RRP into three levels,
the bottom one third or rank 1 refers to bidders are relatively more undervalued, while the top one third or rank 3 refers
to bidders are relatively more overvalued, the middle accounts for the remaining observations. We reported mean value,
t-statistics and the number of the offer premium at each rank. BHAR36ma and BHAR36sa are winsorised at the 1% and
99% level. We performed bootstrap estimation of sampling distribution of BHAR36ma and BHAR36sa at 1000
replications, and report mean value, p-value and the number of BHAR36ma and BHAR36sa of each rank. The mean
difference of t-tests is reported at the end of the table. T-statistics (or p-value) are reported in parentheses. Statistical
significance at the 1%, 5% and 10% levels is denoted ***, ** and * respectively.
Panel A: Market-adjusted BHARs by RRP quartiles
All stock-financed acquisitions Offer Premiums t-stat. N BHAR36ma p-value N
1 (RRP<0) 0.296*** (12.12) 152 -0.348*** (0.000) 137
2 0.251*** (11.74) 152 -0.221** (0.017) 144
3 0.283*** (12.02) 152 -0.198** (0.025) 145
4 (RRP>0) 0.395*** (13.86) 152 -0.161* (0.093) 147
Mean difference 4-1 0.099*** (2.63) 0.187* (0.079)
Panel B: Size-adjusted BHARs by RRP
All stock-financed acquisitions Offer Premiums t-stat. N BHAR36sa p-value N
1 0.294*** (14.53) 201 -0.400*** (0.003) 158
2 0.262*** (13.04) 201 -0.206** (0.040) 180
3 (RRP>0) 0.360*** (15.29) 206 -0.253*** (0.001) 174
Mean difference 3-1 0.066** (2.11) 0.147* (0.073)
Chapter 4. Tables
227
Table 4.12: Endogeneity issues
This table reports the RRP effect on the offer premium by controlling for endogeneity issues. Results from an OLS
regression and a 2SLS regression are presented in this table. The RRP is treated as an endogenous variable, and the market
52 week high, with the bidder and the target 65 week-high treated as instrumental variables. We first obtain the fitted
value from the regression of the RRP on the instrument variables and then replace the RRP with the fitted value. Results
are reported in the “IV” Column. The market 52-week high is defined as the logarithmic term difference between the
highest total market value (CRSP: TOTVAL) over the 335 calendar days ending 30 days prior to the announcement date
and the total market value 30 days prior to the announcement date. The bidder (the target) 65 week-high is 427 calendar
days ending 30 days prior to the announcement date and the stock price 30 days prior to the announcement date. Variable
definitions are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses. Statistical significance at
the 1%, 5% and 10% levels is denoted ***, ** and * respectively, is reported alongside coefficients. The results of the
Hausman test and the over-identifying restrictions test are reported in the lower part of the table.
Offer Premiums OLS IV
Coef. t-stat. Coef. t-stat.
RRP 0.089*** (4.501) 0.085*** (4.853)
Hostile 0.032 (1.563) 0.032 (1.312)
Tender 0.075*** (5.006) 0.075*** (4.518)
Diversification -0.017 (-1.256) -0.017 (-1.280)
Stock -0.003 (-0.184) -0.003 (-0.179)
Cash -0.031* (-1.917) -0.031* (-1.856)
RelativeSize 0.116*** (7.263) 0.116*** (7.719)
Bidder ROA 0.019 (0.269) 0.018 (0.337)
Bidder MTBV -0.000 (-0.010) -0.000 (-0.005)
Bidder lnMV 0.056*** (9.627) 0.056*** (10.282)
Bidder Volatility 0.768 (1.047) 0.744 (1.110)
Target ROA 0.141*** (3.347) 0.140*** (4.545)
Target MTBV -0.001 (-0.878) -0.001 (-0.883)
Target lnMV -0.078*** (-11.703) -0.078*** (-12.472)
Target Volatility -0.315 (-0.537) -0.291 (-0.624)
N 1878 1878
adj. R2 0.169 0.169
Hausman test 0.2170 (p=0.6414)
Over-identifying restrictions
(Sargan test) 26.8164 (p=0.0000)
Chapter 4. Tables
228
Table 4.13: Variables correlation matrix
This table reports pairwise Pearson correlation of the variables used in the regression of offer premiums on the RRP. Variable definitions are as in the notes to Table 4.2.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1.RRP 1.0000
2.Hostile -0.0450 1.0000
3.Tender -0.0199 0.1333 1.0000
4.Diversification -0.0072 -0.0175 0.0245 1.0000
5.Stock 0.0957 -0.1122 -0.2918 -0.0289 1.0000
6.Cash -0.0355 0.0518 0.3323 0.1144 -0.5493 1.0000
7.RelativeSize -0.0939 0.2082 -0.0710 -0.1175 0.0101 -0.2716 1.0000
8.Bidder ROA -0.0898 0.0096 0.0884 0.0969 -0.1875 0.1732 -0.1187 1.0000
9.Bidder MTBV 0.0092 -0.0661 -0.0459 -0.0099 0.1511 -0.0770 -0.0789 0.1448 1.0000
10.Bidder lnMV -0.0465 -0.0500 0.0938 0.0764 -0.1840 0.1811 -0.3759 0.3133 0.2044 1.0000
11.Bidder Volatility 0.0667 -0.0353 -0.0688 -0.0681 0.3333 -0.2574 0.1446 -0.4836 0.1110 -0.4176 1.0000
12.Target ROA -0.2785 0.0618 0.0088 0.0286 -0.0651 0.0106 0.1003 0.2997 -0.0042 0.0912 -0.2870 1.0000
13.Target MTBV -0.1282 -0.0369 -0.0187 0.0093 0.1246 -0.0714 -0.0517 0.0379 0.1840 0.1591 0.0662 0.0426 1.0000
14.Target lnMV -0.2091 0.1085 -0.0051 -0.0844 -0.0961 -0.0757 0.1873 0.1755 0.1153 0.6213 -0.3167 0.3045 0.1913 1.0000
15.Target Volatility 0.3084 -0.1088 -0.0072 0.0382 0.2321 -0.0760 -0.1515 -0.2704 0.0929 -0.2373 0.6290 -0.4696 0.0385 -0.5106 1.0000
Chapter 4. Tables
229
Table 4.14: Robustness check: the RRP effect on the probability of using stocks
This table reports logistic regression results of the probability of using stocks on RRP effect. Dependent variable
is “Stock”, which is a dummy variable, taking a value of 1 if acquisitions are 100% financed with stocks, 0
otherwise. Column (1) reports regression results for all-stock financed acquisitions on the RRP for the acquisition
sample that the bidder is relatively more overvalued than the target or RRP>0, column (2) reports regression
results for all-stock financed acquisitions on RRP for the acquisition sample that the bidder is relatively more
undervalued than the target or RRP<0. We control all other variables the same with Table 4.3. There are 50
observations missing due to multicollinearity problem with the year, so that we are left with 1,742 observations,
1,103 for RRP>0 group, and 639 for RRP<0 group. Variable definitions are as in the notes to Table 4.2. P-value
is reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is denoted ***, ** and *
respectively, is reported alongside marginal effects. We transfer coefficients into marginal effect (ME), evaluated
at the sample means of the independent variables.
Binomial Logistic Regression
Stock
(1) (2)
RRP>0 RRP<0
Coef. p-value ME Coef. p-value ME
RRP 0.831*** (0.003) 0.164 -0.232 (0.611) -0.035
Hostile -0.471 (0.235) -0.093 -1.544** (0.019) -0.233
Tender -2.291*** (0.000) -0.453 -3.799*** (0.000) -0.573
Diversification 0.204 (0.237) 0.040 -0.082 (0.752) -0.012
RelativeSize -0.991*** (0.000) -0.196 0.006 (0.982) 0.001
Bidder lnMV -0.431*** (0.000) -0.085 -0.103 (0.319) -0.016
Bidder MTBV 0.036** (0.031) 0.007 0.074*** (0.000) 0.011
Bidder ROA -0.879 (0.273) -0.174 -2.627** (0.026) -0.396
Target lnMV 0.497*** (0.000) 0.098 0.169 (0.162) 0.025
Target MTBV 0.039* (0.090) 0.008 0.087*** (0.003) 0.013
Target ROA 0.144 (0.704) 0.029 -0.192 (0.803) -0.029
Target Volatility -3.052 (0.606) -0.604 14.092 (0.142) 2.124
Bidder Volatility 33.483*** (0.000) 6.621 36.743*** (0.002) 5.538
Bidder Leverage -0.144 (0.124) -0.028 -0.276*** (0.007) -0.042
Target Leverage -0.151** (0.016) -0.030 -0.099 (0.175) -0.015
Bidder CF/E 0.330 (0.338) 0.065 0.342 (0.428) 0.052
Year Yes Yes Industry Yes Yes Constant -1.226 (0.237) -1.744 (0.171)
N 1103 639
Pseudo R2 0.316 0.350
Chapter 4. Tables
230
Table 4.15: Robustness test: The effect of the RRP on the probability of using stocks: Between payment
methods
This table reports multinomial logistic regression results for Stock versus Cash and Stock versus Mixed on the
RRP by different RRP groups. Specifically, RRP>0 represents bidders that are relatively more overvalued than
their targets, and RRP<0 represents bidders that are relatively more undervalued than their targets. “Stock” refers
to acquisitions that are 100% financed by stocks. “Cash” refers to acquisitions that are 100% financed with cash.
“Mixed” refers to acquisitions that are neither 100% cash financed nor 100% stocks financed. There were 5
missing observations that are defined as “Other” in terms of the method of payment in Thomson One. For both
columns, we control for 15 variables as shown in the specification (4) of Table 4.3. Variable definitions are as in
the notes to Table 4.2. P-value is reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is
denoted ***, **, and * respectively.
(1) (2)
Multinomial Logistic Regression Stock Versus Cash Stock Versus Mix
Coef. p-value Coef. p-value
RRP 0.728*** (0.000) 0.425** (0.017)
Hostile -0.943*** (0.006) -0.516 (0.126)
Tender -3.045*** (0.000) -1.831*** (0.000)
Diversification -0.010 (0.947) 0.262* (0.096)
RelativeSize 0.428 (0.143) -0.852*** (0.000)
Bidder lnMV -0.281*** (0.000) -0.228*** (0.000)
Bidder MTBV 0.058*** (0.001) 0.042*** (0.002)
Bidder ROA -1.768** (0.031) -1.294* (0.077)
Target lnMV 0.471*** (0.000) 0.190** (0.015)
Target MTBV 0.059*** (0.003) 0.050*** (0.007)
Target ROA 0.053 (0.887) 0.324 (0.369)
Target Volatility 2.528 (0.651) 4.395 (0.410)
Bidder Volatility 48.423*** (0.000) 19.221** (0.010)
Bidder Leverage -0.181** (0.024) -0.215*** (0.002)
Target Leverage -0.042 (0.451) -0.156*** (0.001)
Bidder CF/E -0.044 (0.885) 0.361 (0.239)
Year -2.097** (0.018) 0.541 (0.494)
Industry Yes Yes Constant Yes Yes N
Pseudo R2
Chapter 4. Tables
231
Table 4.16: Robustness test: The effect of the RRP on offer premiums: By payment methods
This table reports OLS regression results for the offer premium on the RRP. We divided our M&A sample in three
subsamples according to payment methods. Column (1) reports the regression results for the acquisition sample
that is 100% financed by stocks, column (2) reports the regression results for the acquisition sample that is 100%
financed by cash, column (3) reports the regression results for the acquisition sample that is paid with a mixture
of stocks and cash. There are 5 missing observations that are defined as “Other” in terms of the method of payment
in Thomson One. For both columns, we control for 15 variables as shown in the specification (4) of Table 4.3.
Variable definitions are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses. Statistical
significance at the 1%, 5% and 10% levels is denoted ***, **, and * respectively.
(1) (2) (3)
Offer Premiums Stock Cash Mix
Coef. t-stat. Coef. t-stat. Coef. t-stat.
RRP 0.071** (2.239) 0.113*** (3.674) 0.089** (2.303)
Hostile -0.011 (-0.184) 0.038 (1.107) 0.001 (0.030)
Tender 0.073 (1.053) 0.070*** (3.549) 0.098*** (3.434)
Diversification -0.014 (-0.508) -0.009 (-0.467) -0.030 (-1.243)
RelativeSize 0.176*** (4.397) 0.133*** (5.699) 0.097*** (3.389)
Bidder ROA -0.120 (-1.152) 0.303** (2.301) 0.004 (0.034)
Bidder MTBV -0.000 (-0.042) -0.003 (-1.529) 0.003 (1.414)
Bidder lnMV 0.075*** (5.786) 0.038*** (5.027) 0.064*** (4.762)
Bidder Volatility 0.170 (0.138) -0.673 (-0.505) 1.864 (1.392)
Target ROA 0.223*** (3.516) 0.064 (0.940) 0.073 (0.800)
Target MTBV 0.000 (0.145) -0.004 (-1.142) -0.004 (-1.490)
Target lnMV -0.091*** (-6.627) -0.073*** (-7.326) -0.087*** (-6.206)
Target Volatility 0.541 (0.551) 0.026 (0.034) -2.314* (-1.761)
Constant -0.537*** (-3.799) 0.155 (1.322) 0.020 (0.138)
Year Yes Yes Yes Industry Yes Yes Yes N 608 726 539
adj. R2 0.146 0.223 0.218
Chapter 4. Tables
232
Table 4.18: Robustness test: The effect of the RRP on offer premiums: By deal types
This table reports OLS regression results for the offer premium on the RRP. We divided our M&A sample in two
subsamples according to industry relatedness of the two firms involved. Where we define the “Diversification” to
which the primary two Standard Industry Classification (SIC) codes are different between bidders and targets.
Column (1) reports the regression results for the diversified acquisition sample, column (2) reports the regression
results for the undiversified acquisition sample. For both columns, we control for 15 variables as shown in the
specification (4) of Table 4.3. Variable definitions are as in the notes to Table 4.2. Robustness t-statistics are
reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is denoted ***, **, and *
respectively.
(1) (2)
Offer Premiums Diversified deals Undiversified deals
Coef. t-stat. Coef. t-stat.
RRP 0.127*** (4.066) 0.067** (2.566)
Hostile 0.023 (0.657) 0.034 (1.254)
Tender 0.062** (2.471) 0.090*** (4.555)
Stock -0.016 (-0.500) -0.002 (-0.081)
Cash -0.035 (-1.259) -0.039* (-1.917)
RelativeSize 0.136*** (4.795) 0.115*** (5.611)
Bidder ROA -0.050 (-0.316) 0.019 (0.236)
Bidder MTBV 0.003 (1.277) -0.001 (-0.620)
Bidder lnMV 0.047*** (5.384) 0.066*** (7.924)
Bidder Volatility -0.658 (-0.479) 1.414 (1.561)
Target ROA 0.136** (2.016) 0.138** (2.472)
Target MTBV -0.002 (-0.657) -0.001 (-0.774)
Target lnMV -0.082*** (-7.785) -0.080*** (-8.938)
Target Volatility -0.063 (-0.067) -0.679 (-0.872)
Constant 0.046 (0.386) 0.034 (0.327)
Year Yes Yes Industry Yes Yes
N 702 1176
adj. R2 0.159 0.173
Chapter 4. Tables
233
Table 4.18: Robustness test: The effect of the RRP on offer premiums: By deal choice
This table reports OLS regression results for the offer premium on the RRP. We divided our M&A sample in two
subsamples according to deal choice or whether the deal is defined as tender offer according to Thomson One.
Column (1) reports the regression results for the tender offers, column (2) reports the regression results for mergers.
For both columns, we control for 15 variables as shown in the Specification (4) of Table 4.3. Variable definitions
are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses. Statistical significance at the
1%, 5% and 10% levels is denoted ***, **, and * respectively.
(1) (2)
Offer Premiums Tender Merger
Coef. t-stat. Coef. t-stat.
RRP 0.148*** (4.107) 0.078*** (3.464)
Hostile 0.007 (0.190) 0.041 (1.505)
Diversification -0.052** (-2.044) -0.007 (-0.478)
Stock -0.011 (-0.190) 0.005 (0.282)
Cash -0.076** (-2.578) -0.020 (-1.020)
RelativeSize 0.150*** (4.668) 0.108*** (5.862)
Bidder ROA 0.056 (0.489) -0.007 (-0.091)
Bidder MTBV -0.005 (-1.172) 0.001 (0.440)
Bidder lnMV 0.070*** (5.518) 0.052*** (7.675)
Bidder Volatility 4.638*** (3.126) 0.028 (0.033)
Target ROA 0.066 (0.757) 0.167*** (3.512)
Target MTBV -0.002 (-0.800) -0.001 (-0.543)
Target lnMV -0.081*** (-5.372) -0.077*** (-10.246)
Target Volatility -0.350 (-0.349) -0.232 (-0.344)
Constant 0.325*** (2.764) 0.155 (1.430)
Year Yes Yes Industry Yes Yes
N 380 1498
adj. R2 0.224 0.144
Chapter 4. Tables
234
Table 4.19: Robustness test: The effect of the RRP on offer premiums: By deal completion
This table reports OLS regression results for the offer premium on the RRP. The M&A sample was divided in two
subsamples according to whether the deal is completed during the sample time period studied. Column (1) reports
the regression results for the completed deals or successful deals, column (2) reports the regression results for
deals that are failed to complete or unsuccessful deals. For both columns, we control for 15 variables as shown in
the specification (4) of Table 4.3. Variable definitions are as in the notes to Table 4.2. Robustness t-statistics are
reported in parentheses. Statistical significance at the 1%, 5% and 10% levels is denoted ***, **, and *
respectively.
(1) (2)
Offer Premiums Successful deals Unsuccessful deals
Coef. t-stat. Coef. t-stat.
RRP 0.081*** (3.921) 0.119** (2.347)
Hostile 0.021 (0.517) 0.092*** (2.693)
Tender 0.066*** (4.013) 0.100** (2.560)
Diversification -0.015 (-1.007) 0.022 (0.609)
Stock 0.010 (0.515) -0.017 (-0.375)
Cash -0.033* (-1.842) 0.014 (0.311)
RelativeSize 0.130*** (7.074) 0.134*** (3.422)
Bidder ROA 0.024 (0.320) -0.295 (-1.512)
Bidder MTBV -0.001 (-0.441) 0.005 (1.482)
Bidder lnMV 0.052*** (8.597) 0.102*** (4.315)
Bidder Volatility 0.391 (0.462) 1.543 (1.091)
Target ROA 0.125*** (2.658) 0.291*** (2.830)
Target MTBV -0.001 (-0.534) -0.004 (-0.842)
Target lnMV -0.076*** (-10.891) -0.118*** (-4.747)
Target Volatility -0.263 (-0.402) -0.685 (-0.515)
Constant 0.022 (0.240) -0.158 (-1.015)
Year Yes Yes Industry Yes Yes N 1597 277
adj. R2 0.164 0.238
Chapter 4. Tables
235
Table 4.20: Robustness test: market-adjusted model
This table reports the OLS regression and 2SLS regression results for both bidder and target 3-day CARs on offer
premiums. The dependent variable for specifications (1) and (2) is the bidder CAR3 and for specifications (3) and
(4) is the target CAR3, where CAR3 is calculated with the market-adjusted model. Specifications (2) and (4)
reports the results of 2SLS regression using RRP as an instrument variable of offer premiums. Variable definitions
are as in the notes to Table 4.2. Robustness t-statistics are reported in parentheses. Statistical significance at the
1%, 5% and 10% levels is denoted ***, **and * respectively. We report the results of the Hausman test and F-test
at the lower part of the table.
(1) (2) (3) (4)
Bidder CAR3 Target CAR3
OLS IV OLS IV
Offer Premiums -0.014** -0.070** 0.469*** 0.688***
(-2.167) (-2.078) (20.616) (6.173)
Hostile -0.002 -0.002 -0.002 -0.004
(-0.385) (-0.366) (-0.127) (-0.218)
Tender 0.008** 0.014** 0.040*** 0.017
(2.102) (2.471) (3.022) (0.970)
Diversification 0.003 0.003 0.003 0.003
(0.764) (0.817) (0.305) (0.312)
Stock -0.010** -0.009** -0.046*** -0.046***
(-2.139) (-2.047) (-4.023) (-3.727)
Cash 0.020*** 0.020*** 0.030*** 0.035***
(5.066) (4.504) (2.581) (2.793)
RelativeSize -0.005 -0.005 -0.051*** -0.053***
(-1.207) (-1.485) (-6.259) (-5.780)
Bidder lnMV -0.003*** -0.003***
(-3.693) (-3.074)
Bidder MTBV -0.000 -0.000
(-0.345) (-0.186)
Bidder ROA 0.034* 0.033**
(1.877) (2.550)
Target lnMV -0.002 0.004
(-0.827) (1.001)
Target MTBV -0.002* -0.002
(-1.757) (-1.372)
Target ROA -0.008 -0.011
(-0.298) (-0.557)
Constant 0.013* 0.027** 0.107*** 0.008
(1.666) (2.297) (5.604) (0.153)
N 1878 1878 1878 1878
adj. R2 0.050 . 0.374 .
Hausman test X2=2.9735 (p=0.0848) X2=4.3522 (p=0.0371)
F-test X2=60.2985 (p=0.0000) X2=42.9958 (p=0.0000)
236
CHAPTER 5: MARKET
ANTICIPATION AND MERGER
DECISIONS
Chapter 5. Market Anticipation and Merger Decisions
237
Abstract
This chapter examines the market anticipation effect on M&A decisions and outcomes. Prospect
theory predicts sequential risk-taking actions by decision-makers when facing losses. This
implication was applied in the M&A context, which is a firm’s major corporate investment decision.
The market anticipates managers making investment decisions according to its risk tolerance, hence
maintaining confidence in the firm (i.e. market anticipation effect). Managers follow this anticipation
to rationalise the M&A motive. The magnitude of losses was measured with the bidder reference
price, which is the distance between the firm’s 52-week high and its current stock price, illustrating
the market perception of a firm’s performance. Investors holding bidders’ stocks whose value is in
significant decline relative to the reference price tend to believe risky projects should be available to
the firm and if managers fail to engage in them, this will destroy the market confidence. Market
anticipation increases managerial incentive to perform well in the face of performance decline and
triggers a risk-taking appetiser. The level of riskiness of investment decisions was measured with the
target firm-specific risk. The principal findings in this chapter are presented as follows: it was found
that the riskiness of managerial decisions is anticipated by the market and this explains the choice of
takeover targets. Managers who make decisions based upon market anticipation will receive positive
market reactions. Market-anticipated risks may expose managers to great shareholder monitoring,
pushing them to work hard to justify the M&A motive. Specifically, managers tend to exhibit great
efforts in deal negotiation, reflected in the lower offer premiums they pay for the target, indicating
that managers are cautious in the face of losses. The quality of their decisions according to the
reference point proves to be value creation, reflected in positive long-term performance. Findings of
this chapter suggest that the reference point effect has predictability for target selection.
Chapter 5. Market Anticipation and Merger Decisions
238
5.1. Introduction
“Pain, I have already had occasion to observe, is, in almost all cases, a more pungent sensation than
the opposite and correspondent pleasure. The one, almost always, depresses us much more below the
ordinary, or what may be called the natural state of our happiness, than the other ever raises us
above it.”
Adam Smith (1759: 176-177)
One of the most well-documented phenomena of investors’ behaviour is that people tend to take risks
subsequent to follow losses relative to a reference point, predicted in Kahneman and Tversky’s
prospect theory (1979). According to this, the market should be reluctant to realise losses relative to
a reference point, with the anticipation that future gains can be realised through subsequent firms’
risk-taking decisions. In this chapter, this implication is examined in the context of M&As, which is
a firm’s major investment decisions. Market anticipation was measured with the bidder reference
point, the extent to which a bidder’s current price deviates from its 52-week high price. Evidence that
the market tends to gauge gains and losses relative to a salient piece of price information is also
documented in many studies relating to behavioural finance and M&A literature (Odean, 1998;
Kumer et al., 2015; Schneider and Spalt, 2015)75. Risk magnitude of merger and acquisition (M&A)
decisions was measured with target firm-specific risk similar to that employed by Kumar (2009).76
75 Using the firm’s 52-week high as the reference price is not new. Huddart et al. (2009) identified abnormal trading
volumes around the 52-week high. Burghof and Prothmann (2011) and Li and Yu (2012) found that abnormal trading
volumes around the 52-week high reflects investor attitude towards market news and uncertainty information. Baker et
al. (2012) suggest that bidders assess the target with the target 52-week high. Consistently, the reference price was applied
to the bidder. It reflects the extent to which the loss relative to an aspiration level that the market is unwilling to realise,
which triggers expectations of the firm’s risk-taking decision.
76 The construction of target firm-specific risk is presented in the methodology section of this chapter.
Chapter 5. Market Anticipation and Merger Decisions
239
Whether market anticipation has predictability for corporate investment decisions was investigated,
and the effect of market-anticipated risks on M&A outcomes was examined. The expectation was to
be able to assess whether managers minimise a firm’s value by taking market-anticipated risks.
Explanations regarding why losses trigger managerial willingness of risk-taking are put forward from
the psychological view and provided with empirical evidence relating to the stock market and M&As.
France (1902) explains that people take risks when they encounter losses. He documented a wide
range of historical facts as well as the psychological origin of humans, revealing that the incentive to
take risks is people’s deepest instincts of tasting the joy of life. Further to this, Kahneman and Tversky
(1979) developed a model portraying the relationship between risk and choice, suggesting that people
are loss-aversion and tend to engage in risky decisions in an attempt to compensate losses. Odean
(1998) documents direct evidence in the stock market that investors are reluctant to realise paper
losses, holding the belief that future gains should be generated.
Kumar (2009) suggests investors’ preference for stocks with a lottery feature, which is more
pronounced for those in a lottery-favourable environment and with low incomes. The author noted
that investors are likely to exaggerate a small chance of a large gain. As a result, they choose stocks
with high idiosyncratic volatility, high idiosyncratic skewness and lower stock price (i.e. lottery-like
stocks). Studies suggest that investors who play games of chance are in the hope that their current
wealth will be improved, indicating that investment decisions are motivated by current losses. Barber
and Odean (2008) noted that investors’ trading decisions are based on the focus of their attention.
Both individual and institutional investors will choose eye-grabbing stocks. In this connection, Yu
and Li (2011) and Bhootra and Hur (2013) attribute investors’ attention to the reference point effect
Chapter 5. Market Anticipation and Merger Decisions
240
that trading strategy is driven by a firm’s peak prices. Similarly, Gamble and Johnson (2014) claimed
that investors gauge their losses and gains relative to prior stock returns and increase their portfolio
risks following losses. Kumar et al. (2015) use prior stock market reactions as reference points and
find that losses reflected in prior stock market reactions drive the firm to take on risky targets.
Many additional studies have also documented salient evidence that losses drive a firm’s risk-taking
decisions. Edmans et al. (2012) suggest current underperformance relative to an aspiration level
increases managerial pressure relative to their peers which pushes the manager to take on risky
projects. In a similar vein, Morrow et al. (2007) reported that managers feel great pressure when their
firm’s current performance is far below the market expectations, leading their attempts at value
creation through M&A activities. Kim et al. (2011) documented evidence of desperate managers
whose aim is to grow their firm through M&As when the firm’s performance falls greatly below the
market expectations. Gorton et al. (2009) proposed a theory of mergers relating to firm size, arguing
that firms whose current performance is poor are associated with great pressures, which motivates
managers to start a race for firm size through M&A activities to eliminate threats of being taken
over.77
The magnitude of loss of the firm was measured with the bidder reference price, i.e. a price deviation
of a firm’s current price from its highest stock price occuring in a one-year window preceding the
takeover announcement date. This gives the market insight into how the firm is currently performing
relative to the 52-week high, which is a salient reference price that shapes investors’ minds regarding
77 In support of this, Dickerson et al. (2003) noted that attack, through acquisitions, is the best way to eliminate takeover
threats.
Chapter 5. Market Anticipation and Merger Decisions
241
change in a firm’s profitability. Thus, this measure reflects the magnitude of loss of a firm. Investors
may lose confidence in the firm when this loss is large, inducing their managers to take on risky
projects that could fundamentally change their current wealth status, or blame them for not working
hard. Following Kumar (2009) and Schneider and Splat (2015), the riskiness of the M&A decisions
was measured with target firm-specific risks, which are more easily to be assessed by the market.
Large firm-level risks also expose managers to great shareholder monitoring, motivating the firm to
work hard.
Prior studies have provided supportive evidence as to why a firm tends to make decisions based on a
single piece of information though in truth that they have a full access to the information about their
own firms. Brandenburger and Polak (1996) suggest that the manager is concerned about the market
and tend to make decisions that the market wants to see. Thus, it should be suggested that managers
should make M&A decision according to the market anticipation. Sacheti et al. (2016) argued that
the manager whose decisions follow market anticipation is attempting to avoid the market’s short-
term criticisms. Similarly, Kau et al. (2008) found that firms listen to the market, in that they cancel
investment decisions if there are negative market reactions.78
Using M&As as a platform to explore the reference point effect on major corporate investment
decisions has three main reasons. First, M&A outcomes will alter a firm’s status and change the
wealth of shareholders. Brenner (1983) indicates that a primary motive of risk-taking is to
fundamentally change the relative position in the distribution of wealth. A firm will be blamed for
78 Luo (2005) and Chikh and Filbien (2011), applying this in different markets, found a consistent view with that of Kau
et al. (2008).
Chapter 5. Market Anticipation and Merger Decisions
242
not making sufficient effort for their shareholders when it fails to undertake the investments the
market expects. Secondly, the fact that the shareholder motivates the firm to increase risks is common
in the CEOs’ M&A compensation literature. For instance, Grinstein and Hribar (2004) find risk-
taking managers are compensated with lucrative compensation packages. Graham et al. (2013)
suggest that shareholders are likely to hire managers with risk-taking incentives to avoid costly
incentive pay. M&As allow investigation of whether firms’ major investment decisions made upon
market anticipation receive market recognition. Third, the researcher further explored whether
managers can control risks through skills and efforts, which allows observers to distinguish risk-
taking from stock market gambling, which purely depends on probability (Kumar, 2009).
This chapter has two main motivations. First, it provides novel insights into the debate on managerial
risk-taking incentives. Previous studies have found that risk-taking incentives are driven by
managerial overconfidence or agency problems and yield contradictory results with respect to bidder
performance. Whilst Malmendiar and Tate (2008) claimed that overconfident managers who tolerate
large risks tend to destroy the wealth of shareholders, Croci and Petmezas (2015) indicated that
managerial risk-taking incentive creates value for the firm. Another line of literature links managerial
risk-taking attitudes with CEOs’ M&A compensation. Datta et al. (2001) indicate that high pay-
performance creates managerial value-maximising incentives. However, this view is challenged by
Sanders and Hambrick (2007) who suggest that an increase in risk-taking does not necessarily
enhance firm value. This research investigates the risk-taking motive with reference point theory,
which explains M&A decisions from human nature response (France, 1902). It was expected that a
firm will follow the market anticipation following losses, and managers should rationalise their risk-
Chapter 5. Market Anticipation and Merger Decisions
243
taking incentive to avoid the market criticism. Thus, market perception has a substantial impact on
M&A decisions.
Secondly, only a few papers link the reference point to target risks. For example, Kumar et al. (2015)
use stock market reaction to a prior acquisition as the reference point. The authors find that prior
market reaction loss induces a firm to acquire risky targets. However, how the decision of prior
acquisition influences the current M&A decision remains largely controversial (Billett and Qian, 2008;
Aktas et al., 2009)79. Moreover, Kumar et al. employed target stock volatility as a proxy for M&A
riskiness. However, it measures the total risk of the market and the firm, making it hard to distinguish
the effect of firm-specific risk from the effect of the market-wide risk.80 To combat these issues, in
this chapter, the bidder 52-week high is used as the reference point, which is a more direct measure
for current market reaction. Schneider and Spalt (2015), who applied Kumar’s idea (2009) of
employing the measure for stock riskiness to the target firm, found a negative relationship between
target risks and bidder performance. They ascribe this to a managerial gambling preference. However,
Schneider and Spalt fail to answer whether such risk-taking has a rational motive. If the market
believes that there are risky projects available to a firm while managers do not take any actions, this
will lead to the investors’ belief that managers have a quiet life by entrenching themselves and
destroying the firm’s value (Bertrand and Mullainathan, 2003). On the other hand, when M&A
decisions are made in relative to market anticipation, the market tends to create an image that the firm
79 Aktas et al. (2009) found the learning hypothesis that managers tend to adjust their behaviour according to the previous
acquisition while Billett and Qian (2008) suggest the self-attribution hypothesis that initial success generates
overconfident managers who tend to engage in more acquisitions, which generates lower shareholder wealth. 80 Often the market tends to react more significantly to the firm-level news than the market news. Firm-specific risks can
be driven by any news relating to the firm level. For example, CEO turnovers, or earnings announcement, which are more
sensitive to outside investors.
Chapter 5. Market Anticipation and Merger Decisions
244
is strong, or the firm strives hard to protect shareholders’ wealth, thus enhancing its confidence in the
firm’s prospects. Croci and Petmezas (2015) suggest that managers with risk-taking incentives tend
to select better quality acquisitions. In addition, March and Shapira (1987) indicate that risk-taking is
an essential part of the managerial role. Based on these arguments, bidder performance is expected to
improve when risk-taking is undertaken with market consent.
Using a sample of 2,018 U.S. public acquisitions announced between 1985 and 2014, a positive
relationship between riskiness of M&A decision and the bidder reference point was established,
suggesting an important role of played by market anticipation in corporate investment decisions. The
results of this chapter continue to hold after a series of tests. Further, it was found that bidders’
announcement returns increase with market-anticipated risks, implying that the market will push the
manager to work hard. Further investigation shows that managers who are exposed to observable
risks tend to exhibit greater efforts on deal negotiation and attempting to signal to the market that
they are good managers, which is consistent with Harford et al. (2012) and Bertrand and Mullainathan
(2003).81
Several contributions have been made to previous literature. This is the first paper, to our knowledge,
investigates the reference point effect on managerial risk-taking incentives in the context of M&As.
It is shown that humans’ psychological bias has a direct impact on corporate investment decisions.
This study focuses on the M&A motives of firms who have announced an M&A deal, which allows
making the distinction between the performance of bidders who undertake M&As with market
81 Their studies find that entrenched managers tend to have safe lives and attempt to avoid being monitored, which leads
to overpayment. It is expected that managers making risky decisions will signal to the market that the interests of the
manager and the shareholder are closely linked.
Chapter 5. Market Anticipation and Merger Decisions
245
anticipation and those without, eliminating any concerns that excessive risks are due to managerial
overconfidence and agency problems. It is expected that those M&A motives are followed by
negative market reactions because the market does not anticipate the firm to do so. This chapter
contributes to behavioural finance literature by directly documenting that not only individual
investors but also institutional investors are influenced by the reference point effect while making
decisions in uncertainty. Further, it offers new insights to the existing literature regarding managerial
risk-taking incentives. It was found that market-anticipated risks have a significant impact on M&A
decisions, attributable to one of the important sources of bidder performance improvement.
The remainder of this chapter is organised as follows: Section 2 summaries relevant literature. Section
3 designs hypotheses. Section 4 summarises the data and presents methodologies. Section 5 examines
the reference point effect on M&A decisions, and the relationship between market-anticipated risks
and bidder performance. Section 6 conducts a series of robustness checks. Section 7 concludes the
chapter.
5.2. Literature review
5.2.1. Reference point theory and gambling
France (1902) documents a wide range of interesting facts relating to the history of gambling across
the world and provides experimental evidence explaining the motivation of people engaging in
gambling activities. He concludes that it is human nature to get involved in games of chance and
gambling. People have a gambling spirit and believe in luck. With regard to the human gambling
spirit, France affirms that people tend not to leave the table while a big risky game is in progress. The
human gambling spirit implies that humans naturally explore uncertain events and their attention
Chapter 5. Market Anticipation and Merger Decisions
246
tends to be intense when information is highly uncertain. He explained luck with the fact that
gamblers believe that if they lose in the first half of the game, they will eventually win back in the
second half. This implies that people tend to align games of chance with an exaggerated feeling for
their skills. Such a stimulated mental intention drives people not to think wisely but to rely on the
gambling impulse.
Kahneman and Tversky (1979) developed prospect theory with an S-shape model portraying the
relationship between risk and choice. It suggests that people seek risk when the current status is below
their aspiration level. They are more sensitive to feelings about loss than gains given that the slope in
the loss domain is much steeper than that in the gain. Due to the fact that people are loss-averse, they
have a great incentive to compensate their losses through risk-taking decisions. The model ascribes
the reason for gambling to people’s strong tendency to loss-aversion.
Kumar (2009), following this line of literature, investigates the stock market and finds evidence that
stock-purchasing decisions of individual investors are loss driven. According to the author, individual
investors have a tendency to buy lottery-type stocks. The lottery-type stocks are those with a
combination of high idiosyncratic volatility and skewness, and a low price. When idiosyncratic
volatility is high, investors may overestimate the chance of extreme events occurred in the past would
appear in the future. When idiosyncratic skewness is high, investors may overestimate the probability
of making huge gains by investing less. The lower price reduces the costs of participation, giving
investors an illusion of control. When a panel data set, collected from a large U.S. brokerage house,
was analysed containing trading positions of individual investors between 1991 and 1996, the author
revealed that investors’ preference for lottery-type stocks increased when economic conditions
Chapter 5. Market Anticipation and Merger Decisions
247
became worse or their income was low. Their data also show that low-income investors underperform
the market, indicating that gambling preference is associated with worse performance. Their study
reflects investors’ gambling spirit in investors’ trading strategy in the stock market.
However, it may be argued that the reason for underperformance is not simply the investors’ risk-
taking behaviour since the investors with a lower income could also be less experienced individuals
who are not able to manage the risk of their portfolios in the stock market. In this case, they might
take too many unexpected risks. Borna and Lowery (1987) define this as ‘pure gambling’ in which
people have no control over the outcome of the events but rely heavily on the probability of the
outcome. These authors noted another form of gambling where people’s skills have an impact on the
outcome of gambling,82 implying that acquisitions belong to this form of gambling since managers
with superior skills can better manage the risks.
Rather than Kumar (2009), who investigated the gambling spirit of less experienced investors in the
stock market, Schneider and Spalt (2015) focused on the motivation of gambling for institutional
investors undertaking M&A activities. Following Kumar’s work (2009), Schneider and Spalt
measured target riskiness with its idiosyncratic volatility, which captures the overall risk of the M&A
decisions. Using a sample of public U.S. M&As between 1987 and 2008, the authors found that risky
firms are likely to be acquired, showing the strong gambling preference of bidder managers. Further,
they suggest that bidders who acquire risky targets destroy the wealth of shareholders and the value
of firm reflected in both short and long runs, concluding that the propensity for managerial gambling
82 For example, Borna and Lowery (1987) further illustrate that one’s good knowledge of horse may result in successful
horse-race betting, while rolling a dice is pure gambling in which depends on probability.
Chapter 5. Market Anticipation and Merger Decisions
248
leads to value-destroying deals. However, their results can be interpreted that managers who gamble
by means of M&As do not offer the market a rational motive given the lack of the current status of
the bidder. Based on this argument, it cannot be determined whether a firm listens to the market and
acquire targets accordingly. Investors might falsely believe that the M&A motive is simply because
their managers ‘go with their guts’ and do not coincide with the market reaction, thus bidders receive
negative market reactions.
5.2.2. Reference point theory and risk-taking
Kahneman and Tversky (1979) suggest that people are aggressive when facing losses, as they believe
there is a chance to break even, whilst people are risk-averse when the possible outcome of the
decision is positive. March and Shapira (1987) provided supportive evidence that managers tend to
avoid losses when their performance meets or exceeds market expectations. Their survey summarises
many studies that link managerial risk perspectives and market expectations. It reveals that some
managers interviewed believe that risk-taking is in an essential role of managers and the satisfaction
of success is obtained from the degree of risks taken. While experiencing a decline in performance,
managerial risk-taking incentive tends to be greater. A majority of managers, according to their
surveyed study, believe that their skills can reduce risks.
Studies also provide empirical evidence supporting the view that managers take risks when facing
performance decline. Morrow et al. (2007) investigated firms’ strategies in the situation where a firm
is unable to meet market expectations. Their study suggests that managers whose firms currently
underperform the market are associated with mounting pressures from their investors whose aim is
to achieve the aspired performance levels. This motivates managers to take actions in an attempt to
Chapter 5. Market Anticipation and Merger Decisions
249
turn around the table. Morrow et al.’s data show that acquisitions can create value for the firm,
reasoning that the market expects that the firms to manage risks with better management skills in
integrating existing resources with new resources. They also noted that firms facing a decline in
performance may decline further when taking actions that have no effect on performance, implying
that the firm may perform even worse in the face of a decline in performance when it fails to persuade
the market.
Likewise, Kim et al. (2011) found managers are subject to the pressure of firm’s growth when facing
a decline in stock market performance. The decline in performance indicates a lack of growth for the
firm, which motivates the firm to gain organisational recovery via M&As. By doing so, managers
tend to pay high offer premiums to achieve the deal. Their study interprets their results with reference
point theory of M&A. Kim et al. indicate the driving force of overpayment is because of firms’
tendency to evaluate their current performance relative to their historical performance and their peers’
performance. When they currently perform below these performance levels, they become risk-takers,
paying high offer premiums. However, firms that currently perform well will be aligned with market
expectation that the firm will hit a similar level of growth in the future, and thus managers are not
willing to engage in acquisitions that often destroy the wealth of shareholders.
Edmans et al. (2012) also assert that firms that engage in M&As is because they are under great
pressure from their industry peers. Gorton et al. (2009) proposed a theory of mergers that links the
M&A motive with takeover threat, suggesting a firm that makes a takeover decisions does so to
eliminate takeover threats. This finding is appealing for medium-size firms of the industry in their
sample. The primary reason for firms engaging in unprofitable acquisitions is to eliminate takeover
Chapter 5. Market Anticipation and Merger Decisions
250
threats and thus maintain control of the firm. If a firm’s primary motive is to complete takeovers to
eliminate any pressure or takeover threats from the same industry, they are likely to choose takeovers
that are easier to complete and firms that potentially bring forth higher returns. Following this
argument, Ferreira and Laux (2007) found firms with larger idiosyncratic risk have fewer anti-
takeover defences, thus, implying that firms that lack of takeover resistance are suitable takeover
targets. In addition, those firms have higher levels of private information flow and information about
future earnings in stock prices, which increase attractiveness for those under great pressures.
Earlier studies indicate that firms’ salient historical prices are often reference point prices that the
market uses to assess the firms’ performances. Baker et al. (2012) proposed the target 52-week high
as a reference point price in M&A pricing decisions. Target shareholders are less likely to forsake
control of the firm when an offer price received is significantly lower than the reference point price.
They judge their managerial performance during M&As according to their expectations, implying
that those failing to meet their performance expectations will incur criticism and even face the risk of
shareholder litigation.
Li and Yu (2013) attributed their findings that investors make decisions based on the salient news of
the market to investors’ attention hypothesis, proposing two proxies for investors’ attentions: the
market 52-week high and the historical high. Specifically, the current market price relative to the
market 52-week high is used as a proxy for investors’ under-reaction on the assumption that sporadic
news leads conservative investors to under-react either to good or bad news as they are not fully
incorporated in the market. Whilst the current market price relative to the historical high is used as a
proxy for investors’ over-reaction, reasoning that a series of good or bad news accumulate great deal
Chapter 5. Market Anticipation and Merger Decisions
251
of market attention, investors who make decisions according to the market historical high are over-
react. Li and Yu’s data cover the average index of the Dow Jones between 1928 and 2009 and define
the market 52-week high as a ratio of the current Dow index and its 52-week high, and the historical
high as a ratio of the current Dow index and its historical high. Their findings show that investors
whose attention is based on the historical high are associated with negative market returns in the
future, whilst those whose attention is based on the 52-week high are associated with positive market
returns, in that both over and under-reactions are eventually corrected by the market. Their study
suggests that investors are attention-driven, making stock market predictable.
Bhootra and Hur (2013) suggest the recent effect, indicating that investors tend to have much more
attention on the events occurred in the recent past than in the distant past. They proposed a proxy for
investors’ recent effect based on the timing of the 52-week high, i.e. the number of days since the
date of the 52-week high arrived. Bhootra and Hur’ s study documents the evidence of investors’
attention-driven performance, indicating that firms whose 52-week high occurred in the recent past
outperform those whose 52-week high occurred in the distant past. In particular, Bhootra and Hur’ s
findings show that the top 10% of the stocks whose 52-week high occurred in the recent past
outperform the bottom 10% of the stocks whose 52-week high arrived in the further distance in
monthly average returns by 0.7%. Investors create a strong momentum of profitability for the stocks
that have achieved their best performance recently. On average, profits created by investors’ attention
are twice as large for stocks that attain peaks recently than a long time previously. On the whole, their
study indicates that recent events are likely to associate much more of investors’ emphasis, implying
that a firm whose performance has recently reached its peak might less likely to initiate a takeover
for the reason that takeovers destroy the wealth of the bidder shareholders.
Chapter 5. Market Anticipation and Merger Decisions
252
Many additional studies also suggest the important role of the reference point effect on risk-taking
actions and employ various reference point candidates. Kumar et al. (2015) used the stock market
reactions to a previous acquisition as a reference point for the decision of the current acquisition. The
authors proposed two competing hypotheses relating to the relationship between risk-taking and
choices to examine the decision-making process: reference point effect and the house money effect.
The former indicates that people take risks when facing losses, while the latter suggests that people
take risks when facing gains. Analysing a sample of 823 acquisitions between 1990 and 2006, Kumar
et al. found that firms increase M&A riskiness with both the abnormal dollar gains from the takeover
announcement of a prior acquisition and the abnormal dollar losses, suggesting both theories play a
role in risk-taking. Their data show that loss-aversion driven risk-taking is associated with weak
negative market reactions to the current acquisition, while risk-taking followed by gains has an
insignificant impact on M&A performance. Their study documents evidence that managers take on
risky projects according to market reactions. However, their study measures target riskiness with
target stock return volatility, which includes both the market overall risk and firm-specific risk. This
proxy is problematic, as it cannot be distinguished whether risk-taking is driven by firm-specific or
market overall risks. If the market overall risk is large, investors may believe that risk-taking is
necessary and engaging any projects may as well associate with large risks, making investors feel no
difference to M&As.
Meanwhile, Gamble and Johnson (2014) found evidence suggesting that investors increase their risk
tolerance following losses. Their study focused on individual investors’ trading strategy in the stock
market. Their sample consists of 78,000 households between 1991 and 1996, which were collected
from a large discount brokerage firm in the United States. Investors tend to exit the market when they
Chapter 5. Market Anticipation and Merger Decisions
253
experience large gains or losses in the first six months of the year, whilst those remaining in the
market tend to lower (increase) their risks subsequently follow gains (losses), which is explained by
the disposition effect.
5.2.3. Risk-taking and M&A outcomes
The relationship between managerial risk-taking attitude and value creation is far less clear according
to March and Shapira (1987) who affirm that risk attitudes vary significantly among different
individual managers. Earlier M&A studies provide rather unambiguous results regarding the
relationship between the firm’s risk-taking attitudes and M&A performance. A thread of literature
suggests that a firm’s risk-taking incentive and M&A outcomes are negatively related, stating the
reason for this is agency conflicts between the managers and the shareholders (Grinstein and Hribar,
2004; Sanders and Hambrick, 2007; Graham et al., 2012) or managerial overconfidence (Malmendier
and Tate, 2005; 2008).
Graham et al. (2012) focused on the relationship between managerial risk-taking attitudes and
corporate investment actions and measured a CEO’s risk profile with a series of survey questions.
The authors established that approximately 10% of the top managers interviewed are risk-averse,
whilst a majority of managers display considerable risk-taking attitude and are likely to engage in
value-destroying deals. Moreover, due to the high level of risk tolerance, managers are optimistic,
leading to a great use of short-term debt to finance M&As. Their study also suggests that the firm is
more willing to recruit risk-taking managers on behalf for their decision-making, as recruiting risk-
averse managers make compensations that motivates costly managerial risk-taking incentive. These
findings imply that it is that shareholders expect their managers to be risk-taking but not too optimistic.
Chapter 5. Market Anticipation and Merger Decisions
254
Likewise, Grinstein and Hribar (2004), who investigated the compensation effect on managerial risk-
taking incentive in a sample of 327 M&A deals between 1993 and 1999, noted that 39% of companies
reward their managers who complete the deal. This leads the researcher to believe that the managerial
incentive of value-maximising tends to reduce if the structure of compensation83 motivates the
managers to take reductant risks. It is also noted that managers in pursuit of high compensations
overpay for the targets (Shleifer and Vishny, 1988).
Malmendier and Tate (2008) found overconfident managers are more likely to engage in risky
acquisitions than their non-overconfident counterparts. Their M&A sample consists of 477 large
publicly traded U.S. firms of Forbes 500 from 1980 to 1994. The study measured CEO
overconfidence with option-based measures, showing that overconfident bidders lower the wealth of
shareholders for 75 basis points as opposed to rational bidders. The authors indicate that compared
with non-overconfident CEOs, overconfident CEOs are less likely to create value via M&As.
Specifically, they are more likely to conduct diversifying acquisitions associated with negative
abnormal returns, as overconfident managers overestimate their skills in an unfamiliar area of
industry. In addition, overconfident managers tend to engage in more acquisitions, indicating that
they underestimate the risks of acquisitions.
Other literature finds contrasting evidence that a managerial risk-taking attitude leads to positive
M&A outcomes. Croci and Petmezas (2015) studied the relationship between CEO risk perspectives
and M&A decisions in a sample of 2,056 bidders engaged in 9,789 acquisitions over the period 1997
83 In this regard, Datta et al. (2001) found that managers of high equity-based compensation firms tend to take value-
enhancing M&A projects. Their study shows compensation motivates risk-taking managers to create value for the firm.
Chapter 5. Market Anticipation and Merger Decisions
255
to 2011 and measured managerial risk perspectives with vega and delta. In their study, vega refers to
CEO wealth to stock return volatility while delta refers to CEO wealth to stock price following the
studies of Guay (1999) and Core and Guay (2002). Their findings also show that bidder
announcement returns increase with vega at the 5% significance level, which suggests little evidence
that risks driven by overconfidence destroy the wealth of shareholders. Rather, managers with great
risk-taking incentive enhance the wealth of shareholders. The authors conclude that managers with a
high level of risk-taking tend to select quality deals. Likewise, Gervais et al. (2011) found that
overconfident managers have a high-risk profile, making shareholders’ convexity of the pay-
performance incentive redundant. Their study suggests that overconfidence serves as a solution that
reduces managerial risk aversion, thus saving the firm’s investment costs. According to their study,
shareholders expect their managers to be risk-taking, while the firm performs significantly lower than
their level of expectation and it will reward managerial risk-taking behaviour with compensation. As
such, shareholders might believe that their interests are aligned with the managers if they take on
risky projects on their behalf.
Moeller (2015), who investigated approximately 25,000 companies over a 20-year period that
involved in over 265,000 M&A transactions, also found a positive relationship between risky
strategies and a firm’s performance. He noted that ‘M&A is now a well-exploited strategy. The
opportunities have therefore shrunk and risk-free deals are unlikely to yield the desired results. In
the current environment, portfolio managers therefore have to take risks.’ This indicates that, in
current investment environment, managers can hardly perform well without engaging in risky M&As.
Chapter 5. Market Anticipation and Merger Decisions
256
Harford et al. (2012) found that risk-averse managers generate negative market reactions. Studying a
sample of 1,905 M&As made by U.S. public bidders between 1990 and 2005, the authors find that
entrenched managers who avoid being monitored by block holders tend to use cash instead of stocks
to finance M&As, and moreover they select public targets and overpay for the takeover targets. Their
study compares risk-averse managers to those who make value-destroying deals for fear of losing
their jobs. Bertrand and Mullainathan (2003) proposed the quiet life hypothesis, whereby entrenched
managers are risk-averse.
Many additional studies have attributed the reason for risk-taking actions driving a firm’s positive
performance to the fact that managers are able to take risks that the market anticipates, as the market
tends also to be risk-taking when facing losses. Managers do what the market anticipates can achieve
a better performance, since it creates an insight in the market that the firm is confident in turning
around its performance, and it is more importantly, the interests of the managers and the shareholders
are closely aligned. Certain literature suggests the listening propensity of managers who make
decisions according to market reactions. Brandenburger and Polak (1996) asserted that managers tend
to make decisions that the market thinks it is right rather than their belief that it is for the interests of
the firm, reasoning that ‘the stock market has opinions as to what choices firms should make’. In this
case, managers make decisions according to the market reactions rather than private information they
have. Sacheti et al. (2016) illustrated this with a cricket game by investigating how cricket players
make decisions when facing social pressure. The cricket captains do not make optimal decisions that
can maximise the probability of winning when they are subject to external pressure. Similarly, Sacheti
et al. suggest that managers who make decisions based on the market anticipation are attempting to
avoid criticisms by the market.
Chapter 5. Market Anticipation and Merger Decisions
257
Following this trains of ideas in relevant literature, studies document evidence that managers make
decisions in favour of the market anticipation (Luo, 2005; Kau et al., 2008; Chikh and Filbien, 2011),
as managers only complete the deal associated with positive market reactions. It is generally believed
that a firm is a financing contract between the managers and shareholders, in that the managers are
hired by shareholders, and should make major investment decisions that are highly committed to
completing the transactions. Managers who serve the best interests of their shareholders should award
by the market. On the other hand, managers should exhibit sufficient effort to prove they are good
managers, thus secure their jobs (Lehn and Zhao, 2006). When a firm experiences losses in
performance, it should become more aggressively to persuade the market that they can reverse the
situation. Risky deals attract great deal of market attention, pushing the firm to work diligently.
Therefore, it is expected that one of the positive sources of abnormal returns generated from M&As,
based on the rationale that the manager and the market have a common belief to the firm’s prospects.
This chapter uses the bidder reference point as a proxy for the market anticipation, which is a
deivation of the bidder’s current price from its 52-week high, reflecting the change in market reaction
relative to the bidder’s best performance achieved over the past year prior to the takeover
annoucement. If bidder’s current price is significantly discounted from the reference point price,
shareholders suffer mental losses, which motivate the managers to take more risks. On the other hand,
the role of the reference point effect rationalises managerial risk-taking behaviour. Managers who
conduct M&As according to the level of risks that the market anticipates leads the market to believe
that their interests are closely aligned with those of the managers, as managers are also exposed to
the strict shareholders’ monitoring by taking on risky projects. While the market reacts negatively to
those taking unexpected risks, such as managerial overconfidence or agency problems. It is also
Chapter 5. Market Anticipation and Merger Decisions
258
suggested that the market shows negative reactions to those remaining salient (i.e. entrenched
managers) when they are expected to take an action. This researcher measures riskiness of M&As
with target firm-specific risks in line with the Schneider and Spalt (2015). This research also focuses
on the performance of the firm whose actions are in favour of its shareholders. Not only is the short-
term performance investigated but also the long-term, since it should be possible to observe the
quality of decisions over a longer period of time.
5.3. Hypotheses development
Tversky and Kahneman (1974) indicate a silent piece of information shapes people’s minds and is
used as a reference point for decision-making. Kahneman and Tversky (1979) applied their idea to
the loss-averse tendency, suggesting that mental losses are gauged by the reference point. People
become aggressive when their expectation deviates greatly from the reference point. Odean (1998)
put forward prospect theory in the stock market , documenting evidence that investors are reluctant
to realise loss, thus they retain losers too long in the belief that future gains can be realised through
risky projects. Based on this argument, shareholders holding the firm’s stocks expect the current
losses to be turned round by managers. This author predicts that market reactions to a firm’s current
status relative to the time when the 52-week high arrives are perceived as a loss, which requires the
firm to respond to the market that the firm is striving to achieve profitability. It is generally believed
that M&As are the easiest solution to boost a firm’s performance when it faces a decline in
performance as noted in the work of Morrow et al. (2007) and Kim et al. (2011), however, M&As,
on average, do not benefit bidder shareholders, based on the assumptions that managers take too many
unexpected risks that destroy the wealth of their shareholders (Malmentier and Tate, 2008), or
manager avoid taking expected risks for job security (Bertrand and Mullainathan, 2003; Harford et
Chapter 5. Market Anticipation and Merger Decisions
259
al., 2012). By contrast, when managers are cautious about the M&A motive and the take risks as the
market anticipates, they reveal to the market that there is no misalignment of interests between the
shareholders and the managers. The reference point price reflects the market’s risk-taking profile,
giving managers an insight into how many risks the firm should take to rationalise its M&A motive.
This author measures the loss magnitude of M&A decisions with target firm-specific risks and expect
that they are increased with perceived loss measured by the bidder reference point. Based on the
reference point effect, this research suggests that there exists a level of risks the market is willing for
the firm to take. When the market feels loss, the firm is anticipated to gain by taking on risky projects.
This leads the first testable hypothesis of this chapter,
H1: There is a positive relationship between the bidder reference point and target risks.
The firm under pressures is prone to listen to the market (Luo, 2005; Kau et al., 2008). When engaging
in risky projects, managers show that they are working hard for the shareholders instead of enjoying
safe lives when the firm’s performance is declining (Bertrand and Mullainathan, 2003). Their risky
actions also attract greater market attention, leading to great managerial incentive to making profits
for the shareholders. Barber and Odean (2008) suggest the attention-grabbing hypothesis that
investors tend to buy eye-catching stocks. France (1902) explains this phenomenon as the human
nature to resolve uncertainty. For instance, people may find it difficult to leave the table when risky
bets are placed unless the outcome appears. Therefore, it should be expected that managers so as to
prove themselves as “good managers” will take levels of risks that the market can tolerate and manage
the deal afterwards to gain market recognition.
Chapter 5. Market Anticipation and Merger Decisions
260
H2: Market-anticipated risks are positively associated with bidder announcement returns.
According to Morrow et al. (2007) and Kim et al. (2011), M&A deals are an important way of value
enhancing for managers when the firm experiences a decline to performance. The pressure triggers
managerial risk-taking incentive. In the meanwhile, managers work hard in an attempt to calm down
the market’s panic. They should judge how many risks should be taken based on the market
anticipation measured with the bidder reference point. Unlike overconfident managers who overpay
for the takeover targets, those taking anticipated risks creates the market an image of smart and
diligent managers. By substantiating this market belief, managers reduce offer premiums and
integrate the firm’s resources following a merger. Therefore, this leads the last testable hypothesis of
this chapter that,
H3: Managers making decisions according to market-anticipated risks work hard.
5.4. Data and methodology
5.4.1. Data
A sample of 36,506 U.S. public acquisitions announced between Jan 1, 1985 and December 31, 2014
was collected from Thomson One. This database contains all deal information used in this chapter.
Deals defined as repurchases according to Thomson One were excluded, which was left with a sample
of 13,482 acquisitions. Acquisitions are material decisions to bidders only if their outcomes can
change shareholders’ fundamental wealth. Based on this criteria, bidders were required to acquire at
least 50% of the target shares during the transactions, which yields 11,043 acquisitions. The deal
sample was matched with COMPUSTAT and CRSP. The former database contains the firm’s
Chapter 5. Market Anticipation and Merger Decisions
261
accounting information and the latter one includes all stock price information. Both bidder and target
stock price studied were not a missing value in CRSP, as they are used to calculate target idiosyncratic
volatility, and bidder announcement returns, which yields 5,640 acquisitions. The method of payment
information is not missing, resulting in 5,297 acquisitions. All accounting variables used in the
regressions were required to be available at the fiscal year end prior to the announcement date in
COMPUSTAT. Both bidder and target market value (MV), return-on-asset (ROA), market-to-book
value (M/B), leverage measured with debt-to-equity ratio (D/E) and relative size measured with deal
value divided by the bidder MV, were not with a missing value. It was left with 2,278 acquisitions
after excluding all bidder variables with a missing value, and result in a final sample of 2,018
acquisitions after excluding all target variables with a missing value.
The reference point effect on M&A decisions was studied by controlling a number of deal and bidder
characteristics. Bidder size is expected to be positively related to target risks, as larger bidders tend
to have greater ability to absorb risks. Firm’s growth opportunities were measured with M/B. It is
generally believed that greater opportunities are associated with higher risks. Firm’s profitability was
measured with ROA. It should be expected that the lower ROA bidders tend to be more aggressive
and take more risks than their higher counterparts. Leverage was measured with D/E, reflecting risks
in capital structure. Lower D/E firms are capable in taking more risks than their higher counterparts
(Uysal, 2011). Small targets with lower profitability and growth opportunities tend to be risky, since
they may need more of bidders’ efforts in integration process. At the deal level, the methods of
payment, deal types and deal attitude that have a significant impact on M&A decisions addressed in
the M&A literature were taken into consideration (Croci and Petmezas, 2015; Schneider and Spalt,
2015).
Chapter 5. Market Anticipation and Merger Decisions
262
A summary statistics for the M&A deal sample was presented in Table 5.1. Of 2,018 acquisitions,
773 were all-cash financed acquisitions, 632 were all-stock financed acquisitions, and 581 were
mixed acquisitions (financed with a mixture of cash and stocks).84 There was a relative smaller
proportion of tender offers as opposed to mergers: 530 and 1,488, hostile acquisitions (129) and
diversified acquisitions (661), and there is a larger proportion of successful acquisitions as compared
with unsuccessful acquisitions: 1593 and 398.
A summary statistics for variables used in regressions was presented in Table 5.2, with variable
definitions are as in the note of this table. As seen, bidders are generally larger, with higher
profitability and greater growth opportunities than targets. In public acquisitions, bidders are those
with larger size and greater power than targets, so that they are capable of managing the deals
following a merger. These findings are consistent with previous M&A studies (Moeller et al., 2004;
Rau and Vermaelen, 1998).
In Table 5.3, the whole sample was divided into risky and less risky acquisition subsamples according
to the riskiness level of the target firm, the definition of which is by Kumar (2009). The mean, median
and statistical differences between the two subsamples was presented in this table. Kumar (2009)
suggests that firms with a lower price and higher firm-specific risks display higher levels of risks,
increasing the riskiness of M&A decisions (Schneider and Spalt, 2015). Similarly, risky acquisitions
refer to those with a target whose idiosyncratic volatility is above the median value of the sample,
and whose price is below the median value of the sample (i.e. target price), while less risky
84 The method of payment for 32 acquisitions is defined as “other” according to Thomson One.
Chapter 5. Market Anticipation and Merger Decisions
263
acquisitions refer to the opposite. As seen, bidders engaging in risky acquisitions have a larger
deviation of the current price from the 52-week high price than those engaging in less risky
acquisitions: 0.423 to 0.172, implying that firms perceived as losses are likely to increase risks in the
major investments. Risky acquisitions involve a smaller bidder with lower ROA, growth
opportunities, and leverage. One possible explanation for this is that small bidders are more
vulnerable, which push managers to eliminate threats through risk-taking activity (Gorton et al.,
2009). The statistics also reveal that bidders in the risky acquisition subsample are likely to control
risks by choosing a smaller target than those in the less risky acquisition subsample, based on the
rationale that those smaller targets are less overvalued.
5.4.2. Methodology
5.4.2.1. Target risks and the bidder 52-week high
Target idiosyncratic volatility was measured with standard deviation of residuals computed from a
Fama and French 5-factor model using daily returns from the period of 6 months to 1 month prior to
the announcement date. Fama-French 5-factor returns were collected from the French’s website.
Compared with a 3-factor model, the 5-factor model controls for additional profitability and
investment factors that have an substantial impact on major investment decisions (Fama and French,
2015),85 which is presented as follows:
1 2 3 4 5( )i f m f iR R R R SMB HML RMW CMA (5.1)
85 Schneider and Spalt (2015) used a Fama-French 3-factor model, whereas Kumer (2009) used the 4-factor model to
obtain idiosyncratic volatility. The results obtained in this chapter are similar when calculating target idiosyncratic
volatility with these two models.
Chapter 5. Market Anticipation and Merger Decisions
264
where iR relates to the individual firm returns, fR relates to risk-free returns, relates to returns on
value-weighted (VW) market portfolio, SMB HML RMW CMA relate to returns on diversified
portfolios of small stocks minus big stocks, high B/M stocks minus low B/M stocks, strong
profitability stocks minus weak profitability stocks, and stocks of low investment firms minus stocks
of high investment firms, respectively. i relates to the residuals of the model.
Target idiosyncratic volatility was then calculated with the standard deviation of i , denotes
iiv in
equation (2).
2
1
n
i
iiiv
N
(5.2)
Finally, target risks are a dummy variable, taking value of 1 if the target is above the median value of
its idiosyncratic volatility and below the median value of its price ending 30 days prior to the
announcement date, whereas the price was expressed in a logarithmic term. Kumar (2009) indicates
stocks with high idiosyncratic volatility indicate that extreme returns occurred in the past are highly
likely occur again, and “cheap bets” are more attractive to investors with the risk-taking incentives.
Therefore, M&A decisions are risky if the target displays a higher level of idiosyncratic volatility and
a lower level of price.
The main variable of interest to be assessed is the bidder reference point (BRP), which is constructed
as follows:
Chapter 5. Market Anticipation and Merger Decisions
265
, , 30ln 52 lni i t i tBRP weekhighprice Stockprice (5.3)
where the iBRP is defined as logarithmic term differences between the bidder highest stock price over
335 calendar days ending 30 days prior to the announcement date and the price ending 30 days prior
to the announcement date. Scaling the bidder price on 30 days prior to takeover announcement
eliminates any concerns regarding price information leakage around the announcement date biases
the actual stock price.
5.4.2.2. Short-term method
Bidder announcement returns were calculated with the market model, with parameters were estimated
over a period 261 up to 28 trading days prior to the takeover announcement date and a 3-day event
window is used.
1it mt itR R (5.4)
where itR denotes holding period returns (CRSP: RET) for firm i in the period t,
mtR denotes value-
weighted market returns including dividends (CRSP: VWRETD), it denotes the error term.
5.4.2.3. Long-term method
Following Loughran and Vijh (1997) , the firm’s long-term performance was calculated with the
market-adjusted buy-and-hold abnormal returns (BHARs). It should be expected that the outcome of
decisions to be only assessed in the long term. A firm’s 36-month BHARs were calculated with the
following equation:
Chapter 5. Market Anticipation and Merger Decisions
266
,
1 1
(1 ) (1 )T T
it it index t
t t
BHAR R R
(5.5)
where itR denotes arithmetic returns for firm i on day t. ,index tR denotes the arithmetic return for the
market index on day t. Bootstrapping test was employed to deal with skewness, and the details of this
test was presented in the methodology section of the Chapter 4.
5.4.2.4. Multivariate analysis
Multivariate analysis was used to examine the relevant factors that have impacts on bidder
announcement returns (Draper and Paudyal, 2008), which is presented as follows:
( 1, 1) 1
1
N
i i i i
i
CAR BRP X
(5.6)
( 1, 36) 1
1
N
i i i i
i
BHAR BRP X
(5.7)
where ( 1, 1)CAR relate to bidder cumulative returns of a day prior to and a day after the announcement
date, ( 1, 36)BHAR refer to market-adjusted buy-and-hold abnormal returns up to 36 months following
acquisitions. The main variable to be assessed is the BRP. iX relate to a series of control variables
that have a significant impact on bidder announcement returns in M&A literature (Moeller et al.,
2004; Alexandridis et al., 2013).
Chapter 5. Market Anticipation and Merger Decisions
267
5.5. Empirical results
5.5.1. Does market anticipation affect M&A decisions?
Table 5.4 presents the logistic regression model of the market anticipation on managerial investment
decisions. The results this researcher obtained show a positive relationship between the BRP and
target risks. Specifically, specification (5) shows that the BRP coefficient is positive and significant
at 1% level (coefficient 1.259, p = 0.000). This coefficient was translated into marginal effect, and
presented alongside Specification (5), yielding similar interpretation of OLS regression. It is therefore
interpreted as that for every 10% increase in the BRP increases a 2.2% likelihood of the firm choosing
a risky target rather than a less risky target. The signs of control variables are generally consistent
with the earlier predictions of this chapter. The goodness-of-fit tests show that logistic regression
model has no specification errors (p = 0.890). 86 The principal results continue to hold when
accounting for different sets of control variables, including deal, bidder and target characteristics, as
reported from specifications (1) to (4) respectively, and all regressions control for year and industry
effects.87
The results for specification (5) indicate that the firm makes major corporate investment decisions
according to market-anticipated risks, which is interpreted with the reference point theory of M&As.
Specifically, the market tends to treat the firm 52-week high price as bidder’s potential profitability
and feel mental losses when its current price is significantly deviated from this aspiration level. This
increases their risk tolerance and pushes the firm to take risks. According to what the prospect theory
predicts (Shefrin and Statman, 1985; Odean, 1998), investors holding losers would presumably
86 The null hypothesis posits that the model has no specification error. The test results lead us not to reject the null
hypothesis.
87 Our results are also consistent when controlling for firm-fixed effect.
Chapter 5. Market Anticipation and Merger Decisions
268
believe that the firm can generate future profitability via embarking on risky projects. If the firm
responds to the market correctly, the market confirms the belief that the firm works in favour of
shareholders’ interests. On the other hand, managers are also subject to take more risks especially the
firm’s performance is temporarily depressed (Kim et al., 2013). The reference point enables the firm
to justify their risk-taking behaviour without bearing market’s blame (Sacheti et al., 2016). The firm
will also take risks to maintain the confidence of shareholders. The firm persuades shareholders that
current losses will be realised through decisions the market anticipates. Findings can be also
interpreted as the disciplinary effect of takeover: acquiring risky targets will also expose managers to
shareholder monitoring, which threatens managers’ job security when they perform below the
market’s anticipation (Lehn and Zhao, 2006). By following the market’s anticipation, managers can
release pressure from making major investment decisions.
5.5.2. Do market-anticipated risks create value?
Thus far, the author has examined that M&A decisions are influenced by what the market has
anticipated. In this section, there follows an investigation of the bidder performance whether the firm
takes risks that the market anticipates. There are a variety of sources driving the manager to take
unanticipated risks, such as managerial overconfidence (Malmendier and Tate, 2008) or agency
problems (Grinstein and Hribar, 2004), which may mask the true effect of risks on bidders’ M&A
performance. It is expected that if the firm that takes risks with a rational motive will be rewarded by
the market. So as to disentangle the effect of expected risks from the unexpected risks, an interaction
variable using target risks and the bidder reference point was contructed, with the expectation that
targets risks positively related to bidders’ announcement returns on the condition that the firm takes
risks according to what the market anticipates.
Chapter 5. Market Anticipation and Merger Decisions
269
Table 5.5 presents OLS regressions of bidder announcement returns. Specifications (1) and (2) show
the effects of target risks and the bidder reference point on bidder announcement returns, respectively.
It emerged that either target risks or the BRP alone does not have an impact on bidder announcement
returns. While specification (3) shows a significantly positive relationship between the joint effect of
target risks and the bidder reference point on bidder announcement returns, suggesting that target
risks can predict bidder’s performance when a bidder’s decision is made according to the market
anticipation. In addition, the results indicate that either target risks or the bidder reference point tend
to decrease bidder announcement returns. On the whole, the findings are consistent with the
prediction that firms taking risks observed by the market will receive positive market reactions.
5.5.3. What is the source of bidder performance improvement?
Since the manager is also exposed to the strict shareholders’ monitoring when taking risks that the
market can observe. Managers will attempt to prove themselves as “good managers” in order to avoid
market disappointment (Luo, 2005). According to this, it is believed that managers wish their efforts
to be easily seen by the market. In this case, there is an expectation that managers will exhibit great
efforts on deal negotiation, reflected in offer premiums they pay for the target, in that the offer
premium determines the wealth distribution of the shareholders.
Table 5.6 presents results of OLS regressions of offer premiums. As seen, specification (3) of this
table shows that both target risks and the bidder reference point will increase offer premiums. Bidders
may find targets with greater uncertainty are hard to value or currently underperformed bidders have
less bargaining power and pay hefty offer premiums to target shareholders accordingly. However, the
joint effect of target risks and the bidder reference point decreases offer premiums significantly
Chapter 5. Market Anticipation and Merger Decisions
270
(coefficient -0.0856, p = 0.000). These findings are consistent with our hypothesis suggesting that
managers will have a direct incentive to decrease offer premiums to show negotiation skills or efforts
to the market. The findings of this table suggest that the primary M&A motive of managers to deal
with pressures of a decline in performance is to content their shareholders.
5.5.4. Do market-anticipated risks push the firm to work hard?
It cannot be expected that the market can fully understand the firm’s decisions in the short run, as
uncertainty prevents the market from assessing the quality of decisions until they are gradually
resolved in the long term. This leads the research to investigate the market-anticipated risks effect on
the firm’s long-term performance.
Results of Table 5.7 report that either target risks or the bidder reference point is positively associated
with the long-term performance, presented in specifications (1) and (2). It should be expected that
target risks are influenced by the bidder reference point as the firm makes decisions based on the
market. As such, an interactive term between target risks and the bidder reference point was
constructed, and presented in specification (3). Results show that effects of these two independent
variables on bidder’s long-term performance diminish when including an interaction variable
between target risks and the bidder reference point. The joint effect of target risks and the bidder
reference point increases with bidder’s long-term performance. These findings suggest that each
individual factor partially explains bidder’s long-term performance and has a substantial impact on
each other. The firm gains positive market reactions in the long term when the firm enhances the
confidence of shareholders. Furthermore, it is suggested that managers will be rewarded when their
efforts are recognised.
Chapter 5. Market Anticipation and Merger Decisions
271
5.6. Robustness checks
5.6.1. Various reference point candidates based on firm level.
The robustness checks in this chapter analysed the reference point effect on target risks using a series
of other reference points addressed in previous literature, as presented in Panel A of Table 5.8. First,
the firm’s historical high was used as a reference point, with the expectation that bidder historical
high plays a role in firm’s corporate investment decision-making. A firm’s historical high reflects
investors’ attentions over a longer period of time relative to the 52-week high, and is regarded as a
reference point candidate for conservative investors (Li and Yu, 2012). Results in specification (1)
show a positive relationship between bidder historical high and target risks, suggesting that
conservatism investors also expect the firm to take risks when it faces a decline in performance.
Secondly, a recent ratio was employed as a proxy for investors’ attention, which measures risk-taking
timing for the market and the measure is similar to Bhootra and Hur (2013). The measure is
constructed as 1 minus the number of days since the firm’s 52-week highest stock price arrives
divided by 365 days. Larger of the ratio indicates that the 52-week high arrives more recently. It
should be expected that events occurred in the recent past are more attractive to the market than
occurred in the far distant. If this is the case, the firm should be reluctant to embark on risky
acquisitions when the firm’s best performance recently arrives than in the far distant, since the best
performance arrives more recently will likely to compensate market’s feelings about loss and increase
the attitude of risk aversion. Our results are consistent with the prediction and presented in
specification (2). Thirdly, the prior market reaction was used as a reference point, taking value of 1
if prior market reaction is negative, and 0 if it is positive. If managerial risk-taking is driven by the
market, current market reaction to losses will induce the firm to take on risky projects. Results in
specification (3) show that prior market losses trigger risk-taking behaviour. Finally, it is expected
Chapter 5. Market Anticipation and Merger Decisions
272
that managers are more risk averse than their shareholders as Kumar (2009) suggested. Using prior
ROA relative to current ROA of a firm as a reference point, taking value of 1 if current ROA is lower
than that of the last year, 0 otherwise. It became apparent that ROA has no predictability for risk-
taking, as presented in specification (4). It can be interpreted as follows: managers are insensitive to
current fundamental loss; instead, they make decisions heavily upon market reactions. Overall, it was
found that the bidder 52-week high has the greatest predictability for risk-taking after controlling for
all relevant reference point candidates, as presented in specification (5).
Results of Table 5.9 show other reference points that influence managerial risk-taking incentives.
First, the market 52-week high was used as a reference point candidate. Baker et al. (2012) suggest
that the market returns are an important influential factor for the firm 52-week high. The firm’s overall
risk-profile changes according to the market condition. If the market risk is high, both the market and
the firm may believe that risk-taking is granted. Results of specification (1) show that the market 52-
week high is positively related to target risks, at 10% significance level. Secondly, peer pressure was
used as a proxy, which reflects the firms’ pressure from their industry peers. An increase in peers’
pressure increases a firm’s risk-taking incentive. According to Edmans et al. (2012), managers whose
firms currently underperform, compared with their peers in the same industry, tend to have
considerable pressures that trigger them to take more risks. Following this, a measurement of peer
pressure was constructed using the logarithmic term differences between the bidder 52-week high
and the 52-week high of the industry medians, depicting the extent to which a firm’s performance
below that of the industry’s median, taking value of 1 if the firm performs below the industry 52-
week high median, 0 otherwise. It is expected that managers feel pressures from their industry peers
when their firms perform significantly lower than the industry median. As seen, specification (2)
Chapter 5. Market Anticipation and Merger Decisions
273
shows a positive relationship between peer pressure and target risks. Once again, the significance and
sign of the bidder 52-week high does not change when including both the market 52-week high and
peer pressure in specification (3), suggesting that the bidder 52-week high is a competent reference
point candidate being used to anticipate the firm’s following investment decisions.
5.6.2. Subsample tests
The results are robust in different subsamples by the methods of payment, deal types, whether the
deal is diversified and whether the deal is successful, as presented in Table 5.10. However, it was
noted that the reference point effect is not pronounced among all-stock financed acquisition and
diversified acquisition subsamples. One possible explanation for this is that all-stock financed
acquisitions have those financially constrained bidders whose investment opportunities are limited
(Eckbo et al., 2016), while acquiring diversified targets may create obstacles for shareholders, as it is
hard for the market to judge whether the firm is able to manage in an unfamiliar area of industry.
5.6.3. Alternative proxies for target risks
Following Schneider and Spalt (2015), M&A riskiness was also measured with the target
idiosyncratic risk and conducted with an OLS regression of target idiosyncratic risk on the bidder
reference point. Results of specification (1) of Table 5.11 measures target risks with total volatility,
which is employed by Kumar et al. (2015). Specification (2) measures target risks whose residuals
were obtained from the market model. Specification (3) measures target risks whose residuals were
obtained from Fama-French 4-factor model, which is consistent with the measure of Kumar (2009).
It was evident that there is a positive relationship between the bidder reference point and target risks.
Chapter 5. Market Anticipation and Merger Decisions
274
In addition, the residuals obtained from the market model and the 4-factor model do not vary
significantly when used for target risks.
5.6.4. Risk-taking and M&A performance
Results that the market-anticipated risks taken by the firm increase bidders’ short- and long-term
abnormal returns continue to hold by different estimation models or window periods selected. Table
5.12 reports the relationship between bidder announcement returns and the market-anticipated risks.
The first three specifications measure bidder announcement returns with the market model and uses
a 5-day window while the last three specifications measure bidder announcement returns with the
market-adjusted model with a 3-day window. A positive relationship was found between bidder
announcement returns and the market-anticipated risks, as reflected in specifications (3) and (6)
respectively. Once again, the results show that the firm takes risks according to the market’s
anticipation gain rewards. Table 5.13 shows the relationship between the long-term performance and
the market-anticipated risks. 12- and 24-month windows were used as robustness checks for the 36-
month window of BHARs. The results obtained in this table show that the market-anticipated risks
are positively related with bidders’ long-term performance, which are consistent with the results
reported in Table 5.7.
5.7. Conclusion
In this chapter, this researcher investigates how market anticipation influences firms’ M&A decisions,
and the effect of market-anticipated risks on M&A outcomes. It became evident that perceived loss
measured by the extent to which the deviation of the current price to the firm 52-week high price
triggers managerial risk-taking incentives. Market-anticipated risks lead to positive bidder
Chapter 5. Market Anticipation and Merger Decisions
275
announcement returns, which can be interpreted as managerial great efforts on deal negotiation.
Managers who take risks according to the market risk perspective will increase the managerial
incentive of proving themselves good decision-makers. By doing so, managers work smart and
diligent, reflected in lower offer premiums they pay for the targets and positive market reactions in
the long term.
This chapter contributes to managerial risk-taking incentives and behavioural finance literature. It
was found that the risk-taking preference of individual investors can be transferred to that of
institutional investors through M&As. The market anticipates firm’s next move, and the firm’s market
decision should be rationalised by shareholders. It became apparent that the market anticipation is a
practice for the firm to follow in order to gain positive market reactions. In addition, though it is hard
to explain whether risk-taking incentive gains market recognition, managers taking risks according
to the market anticipation serves one of the important sources for positive market reactions.
Chapter 5. Tables
276
Table 5.1: Summary statistics for M&A sample
This table reports summary statistics for 2018 U.S. domestic public acquisitions announced between 1985 and 2014. The
number N denotes the number of deals per year. The third and fourth columns present mean and median of deal value.
The fifth to the seventh columns present the method of payment. Here “Stock” refers to all-stock acquisitions. “Cash”
refers to all-cash acquisitions. “Mix” refers to acquisitions that neither all-stocks nor all-cash acquisitions. “Tender” refers
to tender offers. “Diversified” refers to diversified deals in which the primary two Standard Industry Classification codes
are different between bidders and targets. “Hostile” refers to hostile deals. “Successful” refers to whether deals are
completed during the sample period, there are 1991 deals with information about deal status.
Year N Deal Value ($ mil) Payment Methods Tender Diversified Hostile Successful
Mean Median Cash Stock Mix Yes No Yes No Yes No Yes No
1985 15 325.85 30.80 6 3 2 9 6 2 13 1 14 6 3
1986 21 225.92 61.20 14 2 3 12 9 9 12 1 20 15 3
1987 33 690.41 66.00 15 7 10 12 21 14 19 5 28 27 4
1988 40 986.62 116.10 18 11 10 20 20 12 28 6 34 23 15
1989 38 209.58 86.95 21 11 6 11 27 14 24 1 37 26 8
1990 29 481.63 21.60 15 8 5 11 18 9 20 3 26 20 8
1991 37 102.29 26.82 9 18 4 13 24 15 22 - 37 26 8
1992 21 136.98 40.26 6 12 2 5 16 8 13 2 19 14 6
1993 45 515.02 93.60 14 16 14 10 35 18 27 5 40 30 14
1994 52 260.96 86.95 13 30 8 10 42 14 38 7 45 37 14
1995 92 572.47 88.67 17 54 19 15 77 36 56 6 86 73 16
1996 89 726.33 157.54 19 44 26 20 69 30 59 6 83 74 15
1997 133 633.15 217.88 23 60 50 24 109 46 87 3 130 111 22
1998 141 1186.73 140.99 35 56 49 31 110 44 97 3 138 126 15
1999 163 1505.82 304.21 58 58 46 45 118 67 96 14 149 131 32
2000 137 2388.55 352.84 30 69 37 31 106 49 88 6 131 113 24
2001 115 1052.84 111.58 35 42 38 26 89 39 76 5 110 96 19
2002 52 1665.40 222.40 22 13 17 14 38 17 35 4 48 47 5
2003 77 755.08 126.23 32 17 28 23 54 19 58 5 72 67 10
2004 72 2678.95 438.18 29 17 26 9 63 22 50 3 69 64 8
2005 75 2535.13 327.64 36 13 26 16 59 26 49 5 70 61 14
2006 88 2444.81 509.29 49 14 25 16 72 31 57 4 84 72 16
2007 67 1319.47 686.22 44 6 17 21 46 17 50 2 65 51 16
2008 63 1990.04 230.27 39 7 16 24 39 13 50 10 53 39 24
2009 53 2912.25 268.89 18 13 20 19 34 17 36 - 53 47 6
2010 61 1719.53 440.71 35 9 14 18 43 14 47 5 56 47 14
2011 48 2334.53 546.40 21 8 18 13 35 12 36 9 39 28 20
2012 43 1258.88 606.27 29 3 11 12 31 19 24 - 43 40 3
2013 54 1703.16 781.04 38 3 11 20 34 13 41 3 51 38 16
2014 64 5140.02 1224.41 33 8 23 20 44 15 49 5 59 44 20
Total 2018 1488.04 210.11 773 632 581 530 1488 661 1357 129 1889 1593 398
Chapter 5. Tables
277
Table 5.2: Summary statistics for variables
This table presents the numbers, means, medians, and standard deviations of variables. Target idiosyncratic volatility is
defined as the standard deviation of the residuals computed from a Fama-French 5 factor model using daily returns from
the period of 6 months to 1 month prior to the announcement date. Target price is target stock price ending 30 days prior
to the announcement date, expressed in logarithmic term. Bidder 52-week high is defined as logarithmic term difference
between the bidder highest stock price over 335 calendar days ending 30 days prior to the announcement date and the
price ending 30 days prior to the announcement date. Bidder 3-day CARs are bidder cumulative abnormal returns 3 days
around the announcement date calculated with market model, whose parameters are estimated using the period of 261
days to 28 days prior to the announcement date. BHAR36ma refers to 36-month market-adjusted buy-and-hold abnormal
returns. Deal characteristics include stock, cash, diversification, hostile, and tender offer which are dummy variables,
taking value of 1 if acquisitions are 100% financed with stocks, 100% financed with cash, diversified merger, hostile
merger, and tender offer, and 0 otherwise. Deal value is raw value taken from Thomson One. “Relative Size” is defined
as deal value divided by bidder MV. Firm characteristics include both bidder and target firm characteristics. MV is defined
as the product of market price and outstanding shares (CRSP: SHROUT*PRC), expressed in logarithmic form. MTBV is
market-to-book value, defined as market equity to book equity, where book equity is total shareholders’ equity (Compustat:
SEQ) plus deferred taxes and investment tax credit (Compustat: TXDITC) minus the preferred stock redemption value
(Compustat: PSTKRV). ROA is return-on-asset ratio, defined as net income (Compustat: NI) divided by total asset
(Compustat: AT). Leverage is measured by debt-to-equity ratio, defined as total long-term debt (Compustat: DITT)
divided by book equity. Other variables are those mainly used in robustness checks. Offer premiums are defined as
logarithmic term difference between offer price and target price ending 30 days prior to the announcement date. Bidder
historical high is defined as logarithmic term difference between bidder highest stock price ending 30 days prior to the
announcement date and the price ending 30 days prior to the announcement date. Recency ratio is defined as 1 minus the
number of days divided by 365 days. Market 52-week high is defined as the logarithmic term difference between the
highest total market value (CRSP: TOTVAL) over the 335 calendar days ending 30 days prior to the announcement date
and the total market value 30 days prior to the announcement date. Peer pressure is defined as the difference between the
bidder 52-week high and the bidder 52-week high of the industry median. Prior market loss is a dummy variable, taking
value of 1 if the firm’s market returns calculated from 365 calendar days before announcement date to 28 calendar days
before takeover announcement date is positive, 0 otherwise. Prior ROA loss is a dummy variable, taking value of 1 if the
firm’s current year’s ROA is less than the past year, 0 otherwise. All accounting variables were at the fiscal year prior to
the announcement date, and continuous variables are winsorised at the 1% and 99% level.
Chapter 5. Tables
278
Table 5.2. (continued)
Variable Mean Median N
Main variables
Target idiosyncratic volatility 0.040 0.033 2018
Target stock price 2.253 2.409 2018
Bidder 52-week high 0.308 0.165 2018
Bidder CAR3 -0.005 -0.005 2018
Bidder BHAR36m -0.075 -0.159 1867
Offer premiums 0.304 0.286 1869
Deal characteristics
Deal value 1488.040 210.113 2018
Cash 0.383 - 2018
Stock 0.313 - 2018
Hostile 0.064 - 2018
Tender 0.263 - 2018
Diversification 0.328 - 2018
Relative size 0.429 0.214 2018
Bidder characteristics
Bidder MV 7.227 7.200 2018
Bidder M/B 3.989 2.536 2018
Bidder ROA 0.018 0.046 2018
Bidder leverage 0.638 0.314 2018
Target characteristics
Target MV 5.173 5.080 2018
Target M/B 2.649 1.767 2018
Target ROA -0.053 0.024 2018
Target leverage 0.622 0.151 2018
Other variables
Bidder historical high 0.919 0.711 2018
Bidder recency ratio 0.564 0.613 2018
Market 52-week high 0.062 0.025 2018
Peer pressure 0.114 -0.022 2018
Prior market loss 0.379 - 2018
Prior ROA loss 0.457 - 2018
Chapter 5. Tables
279
Table 5.3: Risky and less risky acquisition sample
This table presents summary statistics for two subsamples by the median value of target idiosyncratic volatility and target
price ending 30 days prior to the announcement date, expressed in logarithmic term. “Risky acquisitions” refer to targets
that are above the median value of their idiosyncratic volatility and below the median value of their stock price, whereas
“Less risky acquisitions” indicate the opposite. Means and medians of the two subsamples are reported, the number of
observations are reported at the lower part of the table. Median values of dummy variables are omitted as they do not
have any statistical meaning. Variable definitions are as in the notes of Table 5.2. Statistical tests for differences in means,
and medians between the two samples are presented. P-value is reported in parentheses. Statistical significance at the 1%
level, 5% level, 10% level, denoted ***, **, and * respectively.
Variables
(1)
Risky acquisitions
(2)
Less risky acquisitions
Difference (1)-(2)
Mean Median N Mean Median N Mean p-value Median p-value
Bidder 52-week high 0.465 0.308 746 0.216 0.120 1272 0.000 0.000
Bidder CAR3 -0.004 -0.007 746 -0.006 -0.004 1272 0.553 0.765
Bidder BHAR36ma -0.069 -0.280 699 -0.079 -0.114 1168 0.813 0.001
Offer premiums 0.383 0.400 661 0.262 0.258 1208 0.000 0.000
Bidder size 6.132 5.896 746 7.870 7.824 1272 0.000 0.000
Bidder M/B 3.581 2.450 746 4.228 2.607 1272 0.027 0.001
Bidder ROA -0.025 0.039 746 0.043 0.049 1272 0.000 0.000
Bidder leverage 0.506 0.173 746 0.714 0.398 1272 0.001 0.000
Target size 3.609 3.580 746 6.090 5.986 1272 0.000 0.000
Target M/B 1.912 1.229 746 3.082 2.076 1272 0.000 0.000
Target ROA -0.168 -0.045 746 0.015 0.040 1272 0.000 0.000
Target leverage 0.523 0.043 746 0.679 0.241 1272 0.039 0.000
Hostile 0.047 - 746 0.074 - 1272 0.017 -
Tender 0.252 - 746 0.269 - 1272 0.406 -
Diversification 0.362 - 746 0.307 - 1272 0.012 -
Stock 0.389 - 746 0.269 - 1272 0.000 -
Cash 0.355 - 746 0.399 - 1272 0.049 -
Relative size 0.372 0.182 746 0.462 0.234 1272 0.001 0.000
Chapter 5. Tables
280
Table 5.4: Reference point effect on M&A decisions
This table presents results of logistic regressions of target risk on bidder 52-week high. Dependent variable is a
dummy variable, taking value of 1 if targets that are above the median value of their idiosyncratic volatility and
below the median value of their stock price ending 30 days prior to the announcement date. Specifications (1) to
(4) present results with different sets of variables, including deal, bidder and target characteristics respectively,
and specification (5) controls for all of these variables. All regressions include year and industry effect whose
coefficients are omitted. Variable definitions are as in the notes of Table 5.2. P-value is reported in parentheses,
marginal effect for specification (5) is reported alongside coefficients. Goodness-of-fit test is reported at the lower
part of the table. Statistical significance at the 1% level, 5% level, 10% level, denoted ***, **, and * respectively.
Dep. var. Target risks (1) (2) (3) (4) (5)
Bidder 52-week high 1.602*** 1.673*** 1.273*** 1.237*** 1.259*** 0.223 (0.000) (0.000) (0.000) (0.000) (0.000)
Hostile -0.350 -0.221 0.171 0.183 0.032 (0.122) (0.372) (0.567) (0.540)
Tender -0.149 -0.428*** -0.031 0.021 0.004 (0.273) (0.007) (0.865) (0.907)
Diversification 0.225** 0.584*** 0.251 0.175 0.031 (0.045) (0.000) (0.116) (0.292)
Stock 0.066 -0.248 -0.020 0.030 0.005 (0.643) (0.116) (0.920) (0.885)
Cash 0.115 0.009 -0.525*** -0.553*** -0.098 (0.430) (0.958) (0.008) (0.006)
Relative size -0.313*** -1.365*** -0.053 0.150 0.027 (0.004) (0.000) (0.748) (0.461)
Bidder MV -0.559*** 0.107* 0.019 (0.000) (0.062)
Bidder M/B -0.011 -0.007 -0.001 (0.253) (0.528)
Bidder ROA -0.768* -0.034 -0.006 (0.093) (0.954)
Bidder leverage 0.027 -0.006 -0.001 (0.617) (0.937)
Target MV -1.406*** -1.501*** -0.266 (0.000) (0.000)
Target M/B -0.045** -0.043** -0.008 (0.023) (0.031)
Target ROA -2.376*** -2.290*** -0.406 (0.000) (0.000)
Target leverage 0.046 0.046 0.008 (0.356) (0.352)
Year & Industry Y Y Y Y Y
Constant -2.496*** -2.441*** 2.550*** 5.997*** 5.698***
(0.000) (0.000) (0.000) (0.000) (0.000)
N 2018 2018 2018 2018 2018
Pseudo R2 0.162 0.171 0.301 0.517 0.518
Pearson chi (2) X2=1885.56 (p = 0.890)
Chapter 5. Tables
281
Table 5.5: OLS regressions of bidder CARs on market-anticipated risks
This table presents results of OLS regressions of bidder announcement returns on target risk. We include an
interaction variable between target risk and bidder 52-week high in specification (3). Variable definitions are as
in the notes of Table 5.2. P-value is reported in parentheses. Statistical significance at the 1% level, 5% level, and
10% level, denoted ***, ** and * respectively.
Bidder CAR3 (1) (2) (3)
Target risks -0.0024 -0.0148*** (0.553) (0.003)
Bidder 52-week high 0.0044 -0.0180** (0.485) (0.042)
Target risks*Bidder 52-week high 0.0373*** (0.002)
Hostile -0.0104* -0.0103* -0.0106* (0.086) (0.088) (0.078)
Tender 0.0153*** 0.0151*** 0.0151*** (0.000) (0.000) (0.000)
Diversification 0.0010 0.0007 0.0012 (0.778) (0.855) (0.742)
Stock -0.0082* -0.0085* -0.0077 (0.095) (0.081) (0.114)
Cash 0.0298*** 0.0298*** 0.0296*** (0.000) (0.000) (0.000)
Relative size -0.0025 -0.0021 -0.0033 (0.599) (0.652) (0.492)
Bidder MV -0.0055*** -0.0051*** -0.0054*** (0.000) (0.000) (0.000)
Bidder M/B -0.0004 -0.0004 -0.0003 (0.319) (0.299) (0.371)
Bidder ROA 0.0435** 0.0464*** 0.0506*** (0.012) (0.007) (0.003)
Bidder leverage 0.0011 0.0012 0.0010 (0.448) (0.407) (0.502)
Constant 0.0238*** 0.0191** 0.0273*** (0.008) (0.025) (0.003)
N 2018 2018 2018
R2 0.092 0.092 0.098
Chapter 5. Tables
282
Table 5.6: Whether market-anticipated risks increase offer premiums?
This table presents results of OLS regressions of offer premiums on market anticipation driven risks. Offer
premiums are defined as the logarithmic difference between the offer price and the target price 30 days prior to
the takeover announcement date, which is the same as Baker et al. (2012). Bidder 52-week high is the bidder
reference point (BRP), target risks is the dummy variable, taking value of 1 if targets are above the median value
of their idiosyncratic volatility and below the median value of their stock price, 0 otherwise. The first regression
presents results of offer premiums on target risk, the second regression presents results of offer premiums on
bidder 52-week high, and the third regression presents results of offer premiums on market anticipation driven
risks, which is an interaction variable between target risk and bidder 52-week high. Control variables in this panel
are as the same as those in specification (5) of Table 5.4. Variable definitions are as in the notes of Table 5.2. P-
value is reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, **
and * respectively.
Offer premiums (1) (2) (3)
Target risks 0.0615*** 0.0803*** (0.000) (0.000)
Bidder 52-week high 0.0544** 0.0958*** (0.016) (0.000)
Target risks*Bidder 52-week high -0.0856** (0.041)
Hostile 0.0379* 0.0398* 0.0383* (0.076) (0.064) (0.072)
Tender 0.0747*** 0.0727*** 0.0711*** (0.000) (0.000) (0.000)
Diversification -0.0227* -0.0227* -0.0228* (0.094) (0.096) (0.092)
Stock 0.0138 0.0110 0.0087 (0.435) (0.534) (0.622)
Cash -0.0265 -0.0296* -0.0259 (0.104) (0.069) (0.111)
Relative size 0.1499*** 0.1491*** 0.1477*** (0.000) (0.000) (0.000)
Bidder MV 0.0736*** 0.0752*** 0.0736*** (0.000) (0.000) (0.000)
Bidder M/B 0.0013 0.0012 0.0011 (0.346) (0.398) (0.415)
Bidder ROA -0.1001 -0.0892 -0.0951 (0.124) (0.170) (0.149)
Bidder leverage -0.0061 -0.0056 -0.0051 (0.268) (0.318) (0.354)
Target MV -0.0883*** -0.0967*** -0.0874*** (0.000) (0.000) (0.000)
Target M/B -0.0000 -0.0003 -0.0003 (0.986) (0.871) (0.862)
Target ROA 0.1388*** 0.1277*** 0.1397*** (0.001) (0.002) (0.001)
Target leverage -0.0072* -0.0077* -0.0071 (0.098) (0.080) (0.103)
Constant 0.1425*** 0.1834*** 0.1206*** (0.000) (0.000) (0.001)
N 1869 1869 1869
R2 0.163 0.162 0.169
Chapter 5. Tables
283
Table 5.7: Do market-anticipated risks push the firm to work hard?
This table presents results of OLS regression of bidder BHARs on target risk. BHARs are calculated with the
market-adjusted model. We include an interaction variable between target risk and bidder 52-week high in
specification (3). Variable definitions are as in the notes of Table 5.2. P-value is reported in parentheses. Statistical
significance at the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
BHAR36m (1) (2) (3)
Target risks 0.1078** -0.0310 (0.025) (0.621)
Bidder 52-week high 0.1900*** -0.0513 (0.006) (0.491)
Target risks*Bidder 52-week high 0.3571*** (0.004)
Hostile 0.1262 0.1230 0.1200 (0.172) (0.180) (0.188)
Tender 0.0405 0.0229 0.0308 (0.387) (0.622) (0.507)
Diversification -0.0789* -0.0716* -0.0788** (0.050) (0.070) (0.050)
Stock -0.1631*** -0.1730*** -0.1655*** (0.003) (0.002) (0.002)
Cash -0.0016 0.0001 -0.0014 (0.973) (0.998) (0.977)
Relative size 0.0684 0.0446 0.0557 (0.162) (0.351) (0.262)
Bidder MV 0.0220** 0.0164* 0.0232** (0.034) (0.093) (0.023)
Bidder M/B -0.0118*** -0.0120*** -0.0117*** (0.000) (0.000) (0.000)
Bidder ROA 0.5056*** 0.5511*** 0.6084*** (0.001) (0.000) (0.000)
Bidder leverage 0.0139 0.0160 0.0160 (0.394) (0.331) (0.332)
Constant -0.2137** -0.1769* -0.2090** (0.030) (0.053) (0.034)
N 1867 1867 1867
R2 0.038 0.042 0.048
Chapter 5. Tables
284
Table 5.8: Robustness tests: Various reference candidates in relation to the firm level
This table presents results of logistic regressions of target risk on selected reference point candidates. Dependent
variable is a dummy variable, taking value of 1 if targets are above the median values of their idiosyncratic
volatility and below the median value of their stock price ending 30 days prior to the announcement date, 0
otherwise. Control variables are the same as those in specification (5) of Table 5.3. Bidder historical high refers
to the extent to which the current stock price deviated from the firm’s highest stock price, measured with
logarithmic term difference between the firm’s historical high and its current price. We collected all U.S. listed
firms whose stock price between 1958 and 2014 based on data availability. Bidder recency ratio is defined as the
days since the firm’s 52-week high arrives, measured with 1 minus the number of days since the firm’s 52-week
highest stock price arrives divided by 365 days. Prior market loss and prior ROA loss is dummy variables, taking
value of 1 if the firm’s market reaction (ROA) is lower than that of the past year, 0 otherwise. Specification (1)
presents results of target risks on bidder historical high. Specification (2) presents results of target risks on bidder
recency ratio. Specification (3) presents results of target risks on bidder prior market loss. Specification (4)
presents results of target risks on bidder ROA loss. Specification (5) presents results of target risks on bidder 52-
week high (BRP), controlling for those reference point candidates as well as those standard variables highlighted
in the M&A literature. Control variables in each regression are the same as those in specification (5) of Table 5.4.
Variable definitions are as in the notes of Table 5.2. P-value is reported in parentheses. Statistical significance at
the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
Chapter 5. Tables
285
Table 5.8. (continued)
Logit model. Target risks (1) (2) (3) (4) (5)
Bidder historical high 0.434*** 0.197
(0.000) (0.115)
Bidder recency ratio -0.694*** -0.011
(0.002) (0.969)
Prior market loss 0.464*** 0.154
(0.003) (0.411)
Prior ROA loss 0.104 0.075
(0.468) (0.608)
Bidder 52-week high 0.997***
(0.000)
Hostile 0.263 0.225 0.194 0.169 0.218
(0.400) (0.464) (0.525) (0.591) (0.468)
Tender 0.015 0.081 0.092 0.103 0.002
(0.933) (0.655) (0.612) (0.572) (0.992)
Diversification 0.127 0.176 0.167 0.179 0.150
(0.439) (0.282) (0.309) (0.277) (0.364)
Stock 0.088 0.041 0.104 0.078 0.054
(0.669) (0.844) (0.614) (0.703) (0.795)
Cash -0.557*** -0.611*** -0.581*** -0.555*** -0.570***
(0.005) (0.003) (0.004) (0.005) (0.005)
RelativeSize 0.163 0.183 0.195 0.189 0.155
(0.418) (0.349) (0.324) (0.336) (0.448)
Bidder MV 0.105* 0.111* 0.110** 0.092 0.115**
(0.063) (0.050) (0.049) (0.100) (0.047)
Bidder M/B -0.003 -0.004 -0.002 -0.008 -0.003
(0.824) (0.710) (0.864) (0.466) (0.781)
Bidder ROA 0.255 -0.228 -0.194 -0.079 0.154
(0.646) (0.679) (0.725) (0.888) (0.794)
Bidder Leverage -0.014 -0.020 -0.031 -0.024 -0.010
(0.851) (0.794) (0.676) (0.746) (0.898)
Target MV -1.495*** -1.506*** -1.507*** -1.490*** -1.509***
(0.000) (0.000) (0.000) (0.000) (0.000)
Target M/B -0.040** -0.042** -0.045** -0.046** -0.039**
(0.034) (0.038) (0.027) (0.020) (0.044)
Target ROA -2.267*** -2.400*** -2.408*** -2.486*** -2.210***
(0.000) (0.000) (0.000) (0.000) (0.000)
Target Leverage 0.048 0.048 0.055 0.056 0.044
(0.334) (0.347) (0.291) (0.275) (0.378)
Year Y Y Y Y Y
Industry Y Y Y Y Y
Constant 5.697*** 6.431*** 5.842*** 6.086*** 5.467***
(0.000) (0.000) (0.000) (0.000) (0.000)
N 2018 2018 2018 2018 2018
pseudo R2 0.512 0.510 0.509 0.506 0.520
Chapter 5. Tables
286
Table 5.9: Robustness test: Do the market returns or peer pressures play a role in the firm’s risk-taking?
This table presents results of logistic regressions of target risk on other variables that have influence on risk
preferences. Dependent variable is a dummy variable, taking value of 1 if targets are above the median values of
their idiosyncratic volatility and below the median value of their stock price ending 30 days prior to the
announcement date, 0 otherwise. Market 52-week high is defined as logarithmic term difference between the
market index 52-week high and the market index 30 days prior to the takeover announcement date. Peer pressure
reflects the extent to which the individual firm’s 52-week high deviated from the industry 52-week high median.
It is a dummy variable, taking value of 1 if the firm’s 52-week high lower than the industry’s 52-week high median,
0 otherwise. Specification (1) report results of target risks on the market 52-week high. Specification (2) reports
results of target risks on the firm’s peer pressure. Specification (3) report target risks on the bidder reference point
which is the bidder 52-week high, controlling for both the market 52-week high and the industry’s pressure. All
specifications control for a series of standard variables highlighted in the prior M&A literature, which are the same
as those presented in specification (5) of Table 5.4. Variable definitions are as in the notes of Table 5.2. P-value is
reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, ** and *
respectively.
Chapter 5. Tables
287
Table 5.9. (Continued)
Logit model. Target risks (1) (2) (3)
Market 52-week high 2.060* 0.329
(0.086) (0.786)
Peer pressure 0.622*** 0.079
(0.000) (0.700)
Bidder 52-week high 1.167***
(0.000)
Hostile 0.201 0.188 0.188
(0.522) (0.534) (0.529)
Tender 0.107 0.085 0.026
(0.553) (0.638) (0.886)
Diversification 0.191 0.185 0.178
(0.244) (0.264) (0.283)
Stock 0.077 0.022 0.026
(0.706) (0.914) (0.901)
Cash -0.566*** -0.581*** -0.559***
(0.004) (0.004) (0.005)
RelativeSize 0.181 0.199 0.153
(0.352) (0.314) (0.452)
Bidder MV 0.088 0.118** 0.108*
(0.113) (0.040) (0.062)
Bidder M/B -0.007 -0.007 -0.007
(0.520) (0.519) (0.535)
Bidder ROA -0.154 -0.201 -0.051
(0.781) (0.721) (0.931)
Bidder Leverage -0.026 -0.009 -0.006
(0.724) (0.901) (0.935)
Target MV -1.490*** -1.496*** -1.501***
(0.000) (0.000) (0.000)
Target M/B -0.045** -0.045** -0.043**
(0.027) (0.022) (0.032)
Target ROA -2.478*** -2.320*** -2.283***
(0.000) (0.000) (0.000)
Target Leverage 0.052 0.047 0.045
(0.310) (0.342) (0.364)
Year Y Y Y
Industry Y Y Y
Constant 6.136*** 5.651*** 5.669***
(0.000) (0.000) (0.000)
N 2018 2018 2018
pseudo R2 0.507 0.512 0.518
Chapter 5. Tables
288
Table 5.10: Robustness test: Subsample tests
This table presents results of logistic regression of target riskiness on bidder 52-week high by different deal
samples. We partition the whole sample by the method of payment, whether the deal is a tender offer, whether the
deal is diversified, and whether the deal is successful. Control variables are as the same as those in specification
(5) of Table 5.4, and suppressed for the sake of brevity. Variable definitions are as in the notes of Table 5.2. P-
value is reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted ***, **
and * respectively.
Logit model. Target risks Payment methods Tender offer
Cash Stock Mix Yes No
Bidder 52-week high 2.012*** 0.467 2.118*** 2.470*** 0.909*** (0.000) (0.279) (0.000) (0.000) (0.003)
Controls Y Y Y Y Y
Constant 5.981*** 8.711*** 3.877 7.185*** 5.985*** (0.000) (0.000) (0.162) (0.000) (0.000)
N 718 629 556 509 1482
Pseudo R2 0.540 0.544 0.585 0.561 0.534 Diversification Successful
Yes No Yes No
Bidder 52-week high 0.620 1.641*** 1.129*** 2.312***
(0.247) (0.000) (0.000) (0.002)
Controls Y Y Y Y
Constant 5.474*** 6.285*** 6.063*** 7.688***
(0.000) (0.000) (0.000) (0.000)
N 659 1357 1587 395
Pseudo R2 0.520 0.545 0.524 0.606
Chapter 5. Tables
289
Table 5.11: Robustness test: OLS regressions of target idiosyncratic risk on the bidder reference point
This table presents results of OLS regression of target idiosyncratic risks (TIR) on the bidder 52-week high (BRP),
where target idiosyncratic risks are obtained from different estimation model used in the literature. The dependent
variable of the first specification is total volatility of the target, which is the standard deviation of the target’s
stock returns in the period of 1 month and 6 months prior to the takeover announcement date. Target idiosyncratic
risks of specification (2) are calculated with the residuals obtained by the market model (MM_TIR). Target
idiosyncratic risks of specification (3) are calculated with the residuals from Fama-French 4 factor model
(FF4factor_TIR). Variable definitions are as in the notes of Table 5.2. All regressions include the year and industry
dummies. P-value is reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level,
denoted ***, ** and * respectively.
OLS regressions (1) (2) (3)
Total Volatility MM_TIR FF4factor_TIR
BRP 0.009*** 0.010*** 0.010***
(0.000) (0.000) (0.000)
Hostile -0.002* -0.003** -0.003**
(0.086) (0.017) (0.020)
Tender -0.001 -0.001 -0.001
(0.150) (0.121) (0.126)
Diversification 0.000 0.000 0.000
(0.667) (0.674) (0.657)
Stock 0.003*** 0.002** 0.002**
(0.003) (0.050) (0.049)
Cash -0.002* -0.002** -0.003***
(0.059) (0.011) (0.010)
RelativeSize 0.001 0.000 0.001
(0.557) (0.578) (0.564)
Bidder MV 0.001*** 0.001** 0.001**
(0.007) (0.026) (0.026)
Bidder M/B 0.000*** 0.000*** 0.000***
(0.002) (0.001) (0.001)
Bidder ROA -0.013*** -0.010*** -0.010***
(0.000) (0.008) (0.008)
Bidder Leverage 0.000 -0.000 -0.000
(0.988) (0.474) (0.463)
Target MV -0.005*** -0.006*** -0.006***
(0.000) (0.000) (0.000)
Target M/B 0.000*** 0.000* 0.000*
(0.009) (0.079) (0.087)
Target ROA -0.022*** -0.024*** -0.024***
(0.000) (0.000) (0.000)
Target Leverage -0.000 -0.000 -0.000
(0.980) (0.789) (0.812)
Year Y Y Y
Industry Y Y Y
Constant 0.041*** 0.046*** 0.046***
(0.000) (0.000) (0.000)
N 2018 2018 2018
R2 0.610 0.599 0.599
Chapter 5. Tables
290
Table 5.12: Robustness test: OLS regressions of bidder CARs on market-anticipated risks
This table reports the results of OLS regressions of bidder announcement returns on the market’s anticipated risks.
We measure the market’s anticipated risks with an interactive term between target risks and the bidder reference
point (BRP). The first three specifications measure the relation between bidder cumulative abnormal returns on 5
days around the announcement date and calculated with the market model (CAR5mm) while the last three
specifications reflect the relation between bidder cumulative abnormal returns on 3 days around the announcement
date and calculated with the market-adjusted model (CAR3ma). Variable definitions are as in the notes of Table
5.2. P-value is reported in parentheses. Statistical significance at the 1% level, 5% level, and 10% level, denoted
***, ** and * respectively.
OLS regressions (1) (2) (3) (4) (5) (6)
Bidder CAR5mm Bidder CAR3ma
Target risks 0.0015 -0.0134** -0.0028 -0.0111**
(0.740) (0.017) (0.485) (0.026)
BRP 0.0106 -0.0155 -0.0013 -0.0177**
(0.150) (0.115) (0.837) (0.041)
Target risks*BRP 0.0418*** 0.0275**
(0.003) (0.021)
Hostile -0.0120* -0.0120* -0.0123* -0.0109* -0.0109* -0.0111*
(0.080) (0.083) (0.076) (0.070) (0.071) (0.065)
Tender 0.0190*** 0.0184*** 0.0186*** 0.0140*** 0.0142*** 0.0142***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Diversification -0.0003 -0.0004 -0.0002 0.0007 0.0004 0.0008
(0.934) (0.915) (0.968) (0.854) (0.912) (0.824)
Stock -0.0103* -0.0110** -0.0100* -0.0078 -0.0078 -0.0071
(0.062) (0.045) (0.065) (0.108) (0.108) (0.138)
Cash 0.0259*** 0.0259*** 0.0256*** 0.0290*** 0.0291*** 0.0289***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
RelativeSize -0.0069 -0.0073 -0.0079 -0.0032 -0.0026 -0.0035
(0.187) (0.153) (0.127) (0.513) (0.581) (0.468)
Bidder MV -0.0062*** -0.0060*** -0.0061*** -0.0054*** -0.0052*** -0.0054***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Bidder M/B -0.0008* -0.0008* -0.0008* -0.0001 -0.0001 -0.0000
(0.074) (0.066) (0.088) (0.828) (0.823) (0.901)
Bidder ROA 0.0623*** 0.0668*** 0.0721*** 0.0375** 0.0378** 0.0409**
(0.002) (0.001) (0.000) (0.026) (0.024) (0.015)
Bidder Leverage 0.0023 0.0024 0.0022 0.0007 0.0007 0.0005
(0.179) (0.154) (0.193) (0.655) (0.642) (0.728)
Constant 0.0305*** 0.0272*** 0.0332*** 0.0257*** 0.0232*** 0.0294***
(0.002) (0.005) (0.001) (0.004) (0.006) (0.001)
N 2017 2017 2017 2018 2018 2018
R2 0.091 0.092 0.098 0.083 0.083 0.086
Chapter 5. Tables
291
Table 5.13: Robustness test: OLS regressions of BHARs on market-anticipated risks
This table reports the results of OLS regressions of bidder buy-and-hold abnormal returns (BHARs) on the
market’s anticipated risks. BHARs are calculated with the market-adjusted model. Dependent variable of the first
three specifications is BHARs over 12 months after the completion of a merger (BHAR12ma), while dependent
variable of the last three specifications is BHARs over 24 months after the completion of a merger (BHAR24ma).
Variable definitions are as in the notes of Table 5.2. P-value is reported in parentheses. Statistical significance at
the 1% level, 5% level, and 10% level, denoted ***, ** and * respectively.
(1) (2) (3) (4) (5) (6)
BHAR12ma BHAR24ma
Target risks 0.0387 -0.0113 0.0489 -0.0143
(0.149) (0.742) (0.196) (0.767)
BRP 0.0244 -0.0820* 0.0522 -0.0729
(0.525) (0.071) (0.273) (0.224)
Target risks*BRP 0.1565** 0.1843**
(0.032) (0.045)
Hostile -0.0301 -0.0309 -0.0322 0.0571 0.0560 0.0544
(0.497) (0.486) (0.469) (0.348) (0.359) (0.372)
Tender 0.0137 0.0102 0.0138 0.0241 0.0183 0.0225
(0.597) (0.692) (0.592) (0.515) (0.620) (0.543)
Diversification -0.0420* -0.0383* -0.0417* -0.0466 -0.0425 -0.0464
(0.064) (0.088) (0.066) (0.143) (0.177) (0.145)
Stock -0.0404 -0.0411 -0.0378 -0.0931** -0.0954** -0.0915**
(0.188) (0.183) (0.220) (0.024) (0.021) (0.028)
Cash 0.0085 0.0087 0.0080 0.0077 0.0081 0.0073
(0.744) (0.739) (0.758) (0.835) (0.826) (0.842)
RelativeSize 0.0181 0.0099 0.0152 0.0393 0.0287 0.0348
(0.531) (0.724) (0.602) (0.309) (0.441) (0.372)
Bidder MV 0.0091 0.0060 0.0092 0.0195** 0.0161** 0.0198**
(0.114) (0.278) (0.111) (0.015) (0.037) (0.014)
Bidder M/B -0.0039* -0.0039* -0.0037* -0.0122*** -0.0122*** -0.0120***
(0.055) (0.056) (0.065) (0.000) (0.000) (0.000)
Bidder ROA 0.2911*** 0.2888*** 0.3145*** 0.4589*** 0.4651*** 0.4951***
(0.004) (0.003) (0.001) (0.000) (0.000) (0.000)
Bidder Leverage -0.0022 -0.0023 -0.0023 0.0122 0.0125 0.0125
(0.830) (0.821) (0.821) (0.326) (0.321) (0.320)
Constant -0.0886 -0.0561 -0.0724 -0.1835** -0.1517** -0.1700**
(0.126) (0.305) (0.220) (0.021) (0.042) (0.035)
N 1867 1867 1867 1867 1867 1867
R2 0.018 0.018 0.022 0.040 0.040 0.043
292
CHAPTER 6: CONCLUSIONS
Chapter 6. Conclusions
293
6.1. Main findings
This thesis has investigated the reference point effect on M&As. The thesis allows the
researcher to assess investors’ major investment decision-making processes in condition of
risks. The thesis uses the firms’ 52-week highs as reference points since they are frequently
reported by the media that shape investors’ minds, used by M&A participants to assess the
M&A motive and performance. This thesis has contributed to M&A literature by enhancing
this new strand of behavioural finance theory of M&As. It has become evident that firms’ 52-
week highs are as important indications for the firm’s bargaining power, managerial market-
timing, and risk-taking.
The reference point theory has been investigated into many key aspects of M&As in this thesis.
First, the target 52-week high effect was first applied in the U.K. setting, it was found that the
reference point effect reinforces the target firm’s bargaining power in the scenarios where the
takeover market is more competitive and bidders endure extensive information barriers.
Secondly, both the bidder and the target reference prices were revealed to give important
indications of the firm’s valuation to the market and the manager alike. The reference point
theory accommodates implications of Shleifer and Vishny’s misvaluation hypothesis (2003),
suggesting that bidder managers can time the market with the reference points. Thirdly, direct
evidence was produced that the bidder reference price reflects the market risk appetiser, which
is transferrable when bidder managers make M&A decisions according to market-anticipated
risks. Besides, it was established that the reference point theory has strong explanatory value
on many key aspects of M&As.
Chapter 3 has investigated a sample of 451 domestic and 155 cross-border public M&As in the
United Kingdom, affirming that the target reference point plays an essential role in bargaining
Chapter 6. Conclusions
294
power. The results obtained in this chapter are consistent with Baker et al.’s U.S. findings
(2012). It is suggested that the U.K. target is much easier to reinforce the bargaining power
with its reference price, based on the rationale that the market is competitive compared with
the United States. Unlike U.S. firms, whose corporate governance system is manager-oriented,
U.K. firms have a shareholders-oriented corporate governance system where the reference
point effect is stronger since managerial decisions are likely to rely on the reference point to
convince their shareholders that interests of the two sides are closely aligned.
According to these predictions, Chapter 3 has found a positive relationship between the target
reference price and offer premiums, suggesting that the reference point effect is translated into
a strong bargaining power of the target for high M&A offer premiums. Further analyses of the
market reactions to bidders whose payment according to the target reference point show
evidence of overpayment among domestic bidders, whereas there is little evidence of
overpayment among cross-border bidders, suggesting that the market takes foreign bidders’
M&A offer premiums according to the target reference point for granted, unlike those paid by
domestic bidders. Two possible explanations were advisable for this finding. First, domestic
bidders are believed to have a more sophisticated networking than foreign bidders, meaning
that they should have more private information to assess the value of the target firm to reduce
M&A offer premiums, but paying according to the target reference price gives the market an
indication that domestic bidders maybe overconfident. Secondly, it is believed that cross-
border bidders in order to enter a more competitive market to protect their shareholders’ rights
tend to have a relatively weak bargaining position, thus offering a price in favour of the target.
Chapter 3 has made two distinct contributions to the reference point theory of M&As. First,
scenarios where the reference point effect was enhanced were found. The reference point effect
Chapter 6. Conclusions
295
is strong in a more competitive market, as compared the findings of this chapter with those of
the United States. The reference point is pronounced in the case of bidders lacking private
information about the target. Secondly, the reference point theory is used to distinguish the
M&A motives between cross-border and domestic bidders. The market regards domestic
bidders’ offer premiums according to the target reference point as an overpayment, whilst
supporting that of cross-border bidders whose offer premiums rely on the target reference point.
Chapter 4 has thoroughly explored the reference point theory in M&A misvaluation. The
reference point prices of the two firms involved were taken into considerations. A proxy of the
relative reference point (RRP) was constructed to explain how bidders time the market. The
results indicate that stocks are 1.04% more likely to be used for every 10% increase in the RRP,
suggesting that bidders use stocks to dilute overvaluation when there is a sign of it according
to the RRP. The findings also show that for a 10% increase in the RRP lead to an increase of
M&A offer premiums by approximately 0.9%, suggesting that the sign of relative valuation
make takeovers costly. Despite paying according to the RRP incurring negative announcement
returns, offer premiums are positively related to the long-run performance of the stock bidders,
indicating the RRP facilities market-timing.
Chapter 4 has made three contributions to M&A literature. First, it reconciles the misvaluation
hypothesis with the reference point theory of M&As. A new proxy was proposed, based on the
market perception of a firm’s valuation, to explain M&A misvaluation. The idea of using the
RRP for misvaluation eliminates any concerns arising from the biases caused by a firm’s
fundamentals. Secondly, direct evidence is offered showing that not only the market but also
the managers are subject to reference-dependence bias. But they were found to treat the
reference point differently, as the market reacts negatively to bid announcements holding that
Chapter 6. Conclusions
296
bidders who pay according to the RRP are paying too much, whilst managers are concerned
themselves with the interests of the shareholders in the long run. By looking at the RRP, bidders
recognise the relatively more overvalued stocks and in exchange for less negative long-run
performance. Thirdly, the RRP justifies the rationale of payment method choices. Lower M&A
offer premiums are associated with cases where bidders pay with stocks instead of cash for
larger RRP acquisitions, and conversely bidders paying with cash for a lower RRP acquisition
than for a higher RRP acquisition.
Chapter 5 has examined the role of the bidder reference point in the managerial risk-taking
behaviour. One of the most important implications of the reference point is that people have a
strong tendency to avoid loss. This implication was tested in the M&A context and the bidder
reference point was proposed, which represents the extent to which a bidder’s current price
deviates from the bidder 52-week high, capturing the market anticipation for a firm’s
investment decisions. A significantly decline in performance relative to the bidder 52-week
high motivates a great risk appetizer for the market, which also motivates the firm to take on
risky projects so as to recover the loss. The measurement of target risks follows the proxy of
Kumar (2009). It became clear that bidder managers select targets according to the market-
anticipated risks and are rewarded by the market.
Specifically, Chapter 5 has found a positive relationship between the riskiness of an M&A
decision and the bidder reference point, suggesting that the market anticipates the M&A
decision. A positive relationship between bidder announcement returns and the market-
anticipated risks was emerged, showing that bidders listening to the market would justify their
M&A motives. It is also believed that bidders taking anticipated risks expose themselves to
shareholders’ monitoring, pushing them to exhibit sufficient effort observed by the market. The
Chapter 6. Conclusions
297
finding was verified by studying the bidder reference point effect on M&A offer premiums and
the bidder long-term performance. Offer premiums were found to be significantly lower when
M&A decisions are based on the market-anticipated risks, suggesting that managers who link
their interests with those of their shareholders exhibit great efforts in negotiation. Further, the
market-anticipated risks are positively related with long-term performance, suggesting that
managers in the face of loss have a strong incentive to prove themselves good managers to
avoid the risk of being replaced and so they work harder to integrate the firm’s resources.
Chapter 5 has made two distinct contributions to previous literature. First, it uses the reference
point theory to address managerial risk-taking behaviour in the M&A context. Managers take
risks according to the bidder reference point, suggesting that the stock market serves as a risk
transfer mechanism. Secondly, the market-anticipated risks were distinguished from
unanticipated risks (i.e. risks driven by managerial overconfidence and agency problems),
resulting in negative market reactions. Managers undertaking M&As are able to justify their
motives by following the market-anticipated risks, and receive positive market reactions
accordingly. It is clear that decisions made upon the market-anticipated risks improve a firm’s
M&A performance.
On the whole, this thesis has enhanced Baker et al.’s findings (2012) and explored reference
point theory of M&A in a number of ways. The principal findings of this thesis suggest that
the target 52-week high represents a role of the target’s bargaining power. When applying the
reference point theory to the misvaluation hypothesis of M&As, the joint effects of the target
and bidder reference point allow an assessment of the way of managers timing the market.
Despite taking too many unanticipated risks in M&As end up value-destroying to the firm
according to eariler M&A literature, Chapter 5 indicates that managers can rationalise their
Chapter 6. Conclusions
298
risk-taking incentive by focusing on the bidder reference point, and performing this practise
with caution and efforts can improve the firm’s M&A performance.
The main findings of this thesis show that both institutional and individual investors are subject
to reference-dependence bias, suggesting that it is a natural human nature response to make
risky decisions in accordance with the reference point. However, managers and the market may
value a firm differently even with the same information, as one of the main findings suggested
in Chapter 4. Findings in Chapter 3 suggest that M&As are not only a bargaining game between
the management teams of the two firms, but also a trade-off game between the managers and
the shareholders in a firm. It is proposed that managers should always give a thought for the
market when making major investment decisions. As Chapter 5 suggested, M&A decisions
made based on the market’s anticipation convince shareholders that the firm works hard to
protect the value of the firm, suggesting that managers would do whatever the market thinks
right in the face of loss in firm’s performance.
6.2. Implications, limitations and recommendations for future research
Although the research was fruitful in finding results regarding the reference point effect on
M&As, it is subject to a number of limitations, implying some proposals for future research.
First, due to data availability, Chapter 3 has a small fraction of cross-border acquisitions in the
U.K. market. Therefore, the reference point effect should be explored further by extending the
sample into a worldwide cross-border acquisition sample. Further research should take into
account some macro-economic factors, such as total trading volumes, and stock markets of the
two countries whose firms are involved in the cross-border acquisition. Firms in a country with
relatively strong in trading and high in stock market quality are more easily to reinforce the
bargaining power, which is in line with Erel et al. (2012). It remains unexplored whether such
Chapter 6. Conclusions
299
macro-economic factors make bidders to be less likely to focus on the target 52-week high,
given the possibility that target shareholders are also willing to sell the firm to a relatively
strong market regardless of the reference point effect. Meanwhile, the fact that the reference
point effect is enhanced in a setting where the corporate governance system is strong is taken
for granted in this thesis. However, it should be noted that firms’ corporate governance quality
varies significantly, which requires inclusion of some corporate governance related proxies in
the firm level.
Secondly, Chapter 5 has employed a number of reference point candidates highlighted in
previous literature as robustness tests for the bidder 52-week high. The regressions gave a
simple indication that the firm 52-week high is the most suitable reference point effect. It
should be expected that managers consider various reference point candidates, which requires
the researcher to construct a composite index of reference points, including the firm-, the
market-, and the sector-levels. By doing so, one should assign a weight for relevant reference
point candidate, similar to an approach used by Hayward and Hambrick (1997), who
constructed a composite index for the hubris management. In addition, this thesis only looks at
public acquisitions since only publicly-listed firms have available 52-week high. However, it
is worth noting that private acquisitions in which bidders encounter higher levels of information
asymmetry should be a more suitable testing ground for the reference point effect. Although it
is hard to find the reference point in private targets, it is proposed that efforts should be made
to explore the reference point price of the bidder, for example, the unique characteristics of the
bidders have undertaken many private acquisitions in the past.
Finally, Chapter 5 has focused on the risk-taking incentive from the perspective of the market.
The risk tolerance of the manager should also be taken into consideration. The consequence is
Chapter 6. Conclusions
300
the question whether the firm that made a failed deal in the past is likely to take on a risky deal,
as opposed to one that have made a successful deal. In other words, the completion status of
the prior deal may be employed as a reference point for a current M&A decision. Though
Chapter 5 suggests that market-anticipated risks improve a firm’s M&A performance, it
remains unexplored that the market-unanticipated risks effect on the firm’s M&A performance
by including managerial overconfidence and corporate governance related proxies. This
Chapter also draws several implications for practitioners. Managers should be concerned about
stock market signals while making major investment decisions, since the market will inform
the firm how many risks it is willing the firm to take. Managers who do not listen to the market
will be judged as overconfidence or irresponsibility. The researcher shows that taking risks
does not necessarily lead to bad outcomes if that they are justifiable. Though quantifying risk
magnitude is not an easy task according to March and Shapira’s view (1987), this researcher
indicates a possible timing for managerial investment decisions, which is largely explained in
the framework of the reference point theory of M&As.
Overall, the idea of using a firm’s reference points should also be applied in certain areas where
managers have an opportunity to time the market, such as SEOs, equity issuance, and in the
settings where the levels of information asymmetry are high.
301
REFERENCES
References
302
References
Agrawal, A., Jaffe, J.F., 2000. The post-merger performance puzzle: Advances in mergers and acquisitions.
Emerald Group Publishing Limited, pp. 7-41.
Agrawal, A., Jaffe, J.F., 2003. Do takeover targets underperform? Evidence from operating and stock returns.
Journal of Financial and Quantitative Analysis 38, 721-746.
Agrawal, A., Jaffe, J.F., Mandelker, G.N., 1992. The post‐merger performance of acquiring firms: A re‐
examination of an Anomaly. Journal of Finance 47, 1605-1621.
Aktas, N., De Bodt, E., Roll, R., 2009. Learning, hubris and corporate serial acquisitions. Journal of Corporate
Finance 15, 543-561.
Alexandridis, G., Fuller, K.P., Terhaar, L., Travlos, N.G., 2013. Deal size, acquisition premia and shareholder
gains. Journal of Corporate Finance 20, 1-13.
Alexandridis, G., Mavrovitis, C. F., & Travlos, N. G., 2012. How have M&As changed? Evidence from the sixth
merger wave. The European Journal of Finance, 18, 663-688.
Alexandridis, G., Petmezas, D., Travlos, N.G., 2010. Gains from mergers and acquisitions around the world: New
evidence. Financial Management 39, 1671-1695.
Alshwer, A.A., Sibilkov, V., Zaiats, N.S., 2011. Financial constraints and the method of payment in mergers and
acquisitions.
Andrade, G., Mitchell, M., Stafford, E., 2001. New evidence and perspectives on mergers.
Andreou, P.C., Louca, C., Petrou, A.P., 2016. Organizational learning and corporate diversification performance.
Journal of Business Research 69, 3270-3284.
Ang, J.S., Cheng, Y., 2006. Direct evidence on the market‐driven acquisition theory. Journal of Financial
Research 29, 199-216.
Antoniou, A., Arbour, P., Zhao, H., 2008. How much is too much: are merger premiums too high? European
Financial Management 14, 268-287.
Antoniou, A., Petmezas, D., Zhao, H., 2007. Bidder gains and losses of firms involved in many acquisitions.
Journal of Business Finance & Accounting 34, 1221-1244.
Armour, J., Skeel, D.A., 2007. The divergence of US and U.K. takeover regulation. Regulation 30, 08-24.
References
303
Asquith, P., 1983. Merger bids, uncertainty, and stockholder returns. Journal of Financial Economics 11, 51-83.
Asquith, P., Bruner, R.F., Mullins, D.W., 1983. The gains to bidding firms from merger. Journal of Financial
Economics 11, 121-139.
Avery, M., 2013. Literature Review for Local Polynomial Regression. Unpublished manuscript, Department of
Statistics, North Carolina State University, Raleigh, NC.
Baker, M., Pan, X., Wurgler, J., 2012. The effect of reference point prices on mergers and acquisitions. Journal
of Financial Economics 106, 49-71.
Baker, M., Wurgler, J., 2012. Behavioral corporate finance: A current survey. Handbook of the Economics of
Finance, 2, edited by George M. Constantinides, Milton Harris, and Rene M. Stulz. Handbooks in Economics.
New York, 2012.
Barber, B.M., Lyon, J.D., 1997. Detecting long-run abnormal stock returns: The empirical power and specification
of test statistics. Journal of Financial Economics 43, 341-372.
Barber, B.M., Odean, T., 2008. All that glitters: The effect of attention and news on the buying behavior of
individual and institutional investors. Review of Financial Studies 21, 785-818.
Barberis, N., Shleifer, A., Vishny, R., 1998. A model of investor sentiment. Journal of Financial Economics 49,
307-343.
Barberis, N., Xiong, W., 2009. What drives the disposition effect? An analysis of a long-standing preference-
based explanation. Journal of Finance 64, 751-784.
Barberis, N.C., 2013. Thirty years of prospect theory in economics: A review and assessment. Journal of
Economic Perspectives 27, 173-195.
Bargeron, L., Schlingemann, F.P., Stulz, R.M., Zutter, C.J., 2010. Are acquisition premiums lower because of
target CEOs’ conflicts of interest? SSRN Working Paper.
Bena, J., Li, K., 2014. Corporate innovations and mergers and acquisitions. Journal of Finance 69, 1923-1960.
Benartzi, S., Thaler, R.H., 1995. Myopic loss aversion and the equity premium puzzle. Quarterly Journal of
Economics 110, 73-92.
Ben-David, I., Drake, M.S., Roulstone, D.T., 2015. Acquirer valuation and acquisition decisions: Identifying
mispricing using short interest. Journal of Financial and Quantitative Analysis 50, 1-32.
References
304
Berger, P.G., Ofek, E., 1995. Diversification’s effect on firm value. Journal of Financial Economics 37, 39-65.
Berkovitch, E., Khanna, N., 1991. A theory of acquisition markets: Mergers versus tender offers, and golden
parachutes. Review of Financial Studies 4, 149-174.
Berkovitch, E., Narayanan, M., 1993. Motives for takeovers: An empirical investigation. Journal of Financial
and Quantitative Analysis 28, 347-362.
Bertrand, M., Mullainathan, S., 2003. Enjoying the quiet life? Corporate governance and managerial preferences.
Journal of Political Economy 111, 1043-1075.
Betton, S., Eckbo, B.E., Thorburn, K.S., 2008. Markup pricing revisited. Unpublished working paper, Tuck
School of Business at Dartmouth, Hanover, NH (2008-45).
Betton, S., Eckbo, B.E., Thorburn, K.S., 2009. Merger negotiations and the toehold puzzle. Journal of Financial
Economics 91, 158-178.
Bhootra, A., Hur, J., 2013. The timing of 52-week high price and momentum. Journal of Banking & Finance 37,
3773-3782.
Billett, M.T., Qian, Y., 2008. Are overconfident CEOs born or made? Evidence of self-attribution bias from
frequent acquirers. Management Science 54, 1037-1051.
Boone, A.L., Mulherin, J.H., 2007. How are firms sold? Journal of Finance 62, 847-875.
Borna, S., Lowry, J., 1987. Gambling and speculation. Journal of Business Ethics 6, 219-224.
Bouwman, C., Fuller, K., Nain, A., 2003. The performance of stock-price driven acquisitions. Social Science
Research Network. SSRN Working Paper.
Bouwman, C.H., Fuller, K., Nain, A.S., 2009. Market valuation and acquisition quality: Empirical evidence.
Review of Financial Studies 22, 633-679.
Bradley, M., Desai, A., Kim, E.H., 1983. The rationale behind interfirm tender offers: Information or synergy?
Journal of Financial Economics 11, 183-206.
Bradley, M., Desai, A., Kim, E.H., 1988. Synergistic gains from corporate acquisitions and their division between
the stockholders of target and acquiring firms. Journal of Financial Economics 21, 3-40.
References
305
Bradley, M., Sundaram, A., 2004. Do acquisitions drive performance or does performance drive acquisitions.
Social Science Research Network.
Brandenburger, A., Polak, B., 1996. When managers cover their posteriors: Making the decisions the market
wants to see. The RAND Journal of Economics 27, 523-541.
Brar, G., Giamouridis, D., Liodakis, M., 2009. Predicting European takeover targets. European Financial
Management 15, 430-450.
Brenner, R., 1983. History-The human gamble. University of Chicago Press: 1st Edition.
Brown, S.J., Warner, J.B., 1985. Using daily stock returns: The case of event studies. Journal of Financial
Economics 14, 3-31.
Burch, T.R., Nanda, V., Silveri, S., 2012. Do institutions prefer high-value acquires? An analysis of trading in
stock-financed acquisitions. Journal of Financial Research 35, 211-241.
Burghof, H.-P., Prothmann, F., 2011. The 52-week high strategy and information uncertainty. Financial Markets
and Portfolio Management 25, 345-378.
Chang, S., 1998. Takeovers of privately held targets, methods of payment, and bidder returns. Journal of Finance
53, 773-784.
Chikh, S., Filbien, J.-Y., 2011. Acquisitions and CEO power: Evidence from French networks. Journal of
Corporate Finance 17, 1221-1236.
Chira, I., Madura, J., 2015. Reference point theory and pursuit of deals. Financial Review 50, 275-300.
Clark, K., Ofek, E., 1994. Mergers as a means of restructuring distressed firms: An empirical investigation.
Journal of Financial and Quantitative Analysis 29, 541-565.
Comment, R., Jarrell, G.A., 1995. Corporate focus and stock returns. Journal of Financial Economics 37, 67-87.
Core, J., Guay, W., 2002. Estimating the value of employee stock option portfolios and their sensitivities to price
and volatility. Journal of Accounting Research 40, 613-630.
Cornu, P., 2000. The deterring role of the medium of payment in takeover contests: Theory and evidence from
the U.K.. European Financial Management 6, 423-440.
References
306
Croci, E., Petmezas, D., 2015. Do risk-taking incentives induce CEOs to invest? Evidence from acquisitions.
Journal of Corporate Finance 32, 1-23.
Croci, E., Petmezas, D., Travlos, N., 2012. Asymmetric information and target firm returns. European Journal of
Finance 18, 639-661.
Croci, E., Petmezas, D., Vagenas-Nanos, E., 2010. Managerial overconfidence in high and low valuation markets
and gains to acquisitions. International Review of Financial Analysis 19, 368-378.
Danbolt, J., 1995. An analysis of gains and losses to shareholders of foreign bidding companies engaged in cross-
border acquisitions into the United Kingdom, 1986–1991. European Journal of Finance 1, 279-309.
Danbolt, J., 2004. Target company Cross-border effects in acquisitions into the U.K.. European Financial
Management 10, 83-108.
Datta, D.K., Pinches, G.E., Narayanan, V., 1992. Factors influencing wealth creation from mergers and
acquisitions: A meta‐analysis. Strategic Management Journal 13, 67-84.
Datta, S., Iskandar-Datta, M., Raman, K., 2001. Executive compensation and corporate acquisition decisions.
Journal of Finance 56, 2299-2336.
De, S., Jindra, J., 2012. Why newly listed firms become acquisition targets. Journal of Banking & Finance 36,
2616-2631.
DeBondt, W.F., Thaler, R., 1985. Does the stock market overreact? Journal of Finance 40, 793-805.
Denis, D.J., Denis, D.K., Yost, K., 2002. Global diversification, industrial diversification, and firm value.
Journal of Finance 57, 1951-1979.
Devos, E., Kadapakkam, P.R., Krishnamurthy, S., 2009. How do mergers create value? A comparison of taxes,
market power, and efficiency improvements as explanations for synergies. Review of Financial Studies 22, 1179-
1211.
Di Giuli, A., 2013. The effect of stock misvaluation and investment opportunities on the method of payment in
mergers. Journal of Corporate Finance 21, 196-215.
Díaz, K.B.D., Azofra, S.S., Gutiérrez, C.L., 2009. Are M&A premiums too high? Analysis of a quadratic
relationship between premiums and returns. Quarterly Journal of Finance and Accounting 48, 5-21.
References
307
Dickerson, A.P., Gibson, H.D., Tsakalotos, E., 2003. Is attack the best form of defence? A competing risks
analysis of acquisition activity in the U.K.. Cambridge Journal of Economics 27, 337-357.
Dodd, P., 1980. Merger proposals, management discretion and stockholder wealth. Journal of Financial
Economics 8, 105-137.
Dodd, P., Ruback, R., 1977. Tender offers and stockholder returns: An empirical analysis. Journal of Financial
Economics 5, 351-373.
Dong, M., Hirshleifer, D., Richardson, S., Teoh, S.H., 2006. Does investor misvaluation drive the takeover market?
Journal of Finance 61, 725-762.
Doukas, J., 1995. Overinvestment, Tobin's q and gains from foreign acquisitions. Journal of Banking & Finance
19, 1285-1303.
Doukas, J., Travlos, N.G., 1988. The effect of corporate multinationalism on shareholders’ wealth: Evidence from
international acquisitions. Journal of Finance 43, 1161-1175.
Doukas, J.A., Petmezas, D., 2007. Acquisitions, overconfident managers and self‐attribution bias. European
Financial Management 13, 531-577.
Draper, P., Paudyal, K., 2006. Acquisitions: private versus public. European Financial Management 12, 57-80.
Draper, P., Paudyal, K., 2008. Information asymmetry and bidders’ gains. Journal of Business Finance &
Accounting 35, 376-405.
Eckbo, B.E., 2009. Bidding strategies and takeover premiums: A review. Journal of Corporate Finance 15, 149-
178.
Eckbo, B.E., 2014. Corporate takeovers and economic efficiency. Annual Review of Financial Economics 6, 51-
74.
Eckbo, B.E., Langohr, H., 1989. Information disclosure, method of payment, and takeover premiums: Public and
private tender offers in France. Journal of Financial Economics 24, 363-403.
Eckbo, B.E., Makaew, T., Thorburn, K.S., 2016. Are stock-financed takeovers opportunistic? SSRN Working
Paper.
Eckbo, B.E., Thorburn, K.S., 2000. Gains to bidder firms revisited: domestic and foreign acquisitions in Canada.
Journal of Financial and Quantitative Analysis 35, 1-25.
References
308
Edmans, A., Goldstein, I., Jiang, W., 2012. The real effects of financial markets: The impact of prices on takeovers.
Journal of Finance 67, 933-971.
Edmister, R., Walkling, R., 1985. Determinants of tender offer premiums. Financial Analysts Journal Jan.-Feb.,
27-37.
Edwards, W., 1968. Conservatism in Human Information Processing. In B. Kleinmuntz (Ed.), Formal
Representation of Human Judgment (pp. 17-52). New York: Wiley.
Erel, I., Liao, R.C., Weisbach, M.S., 2012. Determinants of cross‐border mergers and acquisitions. Journal of
Finance 67, 1045-1082.
Faccio, M., Masulis, R.W., 2005. The choice of payment method in European mergers and acquisitions. Journal
of Finance 60, 1345-1388.
Falaato, A., 2008. Superstars or superlemons? Top executive pay and corporate acquisitions, American Law &
Economics Association Annual Meetings: bepress, p. 102.
Fama, E., Fisher, L., Jensen, M., Roll, R., 1969. The adjustment of stock prices to new information. International
Economic Review 10.
Fama, E.F., 1998. Market efficiency, long-term returns, and behavioral finance. Journal of Financial Economics
49, 283-306.
Fama, E.F., French, K.R., 1993. Common risk factors in the returns on stocks and bonds. Journal of Financial
Economics 33, 3-56.
Fama, E.F., French, K.R., 1996. Multifactor explanations of asset pricing anomalies. Journal of Finance 51, 55-
84.
Fama, E.F., French, K.R., 2015. A five-factor asset pricing model. Journal of Financial Economics 116, 1-22.
Ferreira, M.A., Laux, P.A., 2007. Corporate governance, idiosyncratic risk, and information flow. Journal of
Finance 62, 951-989.
Fishman, M.J., 1989. Preemptive bidding and the role of the medium of exchange in acquisitions. Journal of
Finance 44, 41-57.
France, C.J., 1902. The gambling impulse. American Journal of Psychology 13, 364-407.
References
309
Franks, J., Harris, R., Titman, S., 1991. The post-merger share-price performance of acquiring firms. Journal of
Financial Economics 29, 81-96.
Franks, J.R., Broyles, J.E., Hecht, M.J., 1977. An industry study of the profitability of mergers in the United
Kingdom. Journal of Finance 32, 1513-1525.
Franks, J.R., Harris, R.S., 1989. Shareholder wealth effects of corporate takeovers: the U.K. experience 1955–
1985. Journal of Financial Economics 23, 225-249.
Franks, J.R., Harris, R.S., Mayer, C., 1988. Means of payment in takeovers: Results for the United Kingdom and
the United States, Corporate takeovers: Causes and consequences: University of Chicago Press, pp. 221-264.
Fu, F., Lin, L., & Officer, M. S., 2013. Acquisitions driven by stock overvaluation: Are they good deals? Journal
of Financial Economics, 109, 24-39.
Fuller, K., Netter, J., Stegemoller, M., 2002. What do returns to acquiring firms tell us? Evidence from firms that
make many acquisitions. Journal of Finance 57, 1763-1793.
Galasso, A., Simcoe, T.S., 2011. CEO overconfidence and innovation. Management Science 57, 1469-1484.
Gamble, K.J. and Johnson, B., 2014. How Prior Outcomes Affect Individual Investors’ Subsequent Risk Taking.
SSRN Working Paper.
George, T.J., Hwang, C.Y., 2004. The 52‐week high and momentum investing. Journal of Finance 59, 2145-2176.
Gervais, S., Heaton, J.B., Odean, T., 2011. Overconfidence, compensation contracts, and capital budgeting.
Journal of Finance 66, 1735-1777.
Ghosh, A., 2001. Does operating performance really improve following corporate acquisitions? Journal of
Corporate Finance 7, 151-178.
Golubov, A., Petmezas, D., Travlos, N.G., 2012. When it pays to pay your investment banker: New evidence on
the role of financial advisors in M&As. Journal of Finance 67, 271-311.
Golubov, A., Petmezas, D., Travlos, N.G., 2015. Do stock-financed acquisitions destroy value? New methods and
evidence. Review of Finance, rfv009.
Gorton, G., Kahl, M., Rosen, R.J., 2009. Eat or be eaten: A theory of mergers and firm size. Journal of Finance
64, 1291-1344.
References
310
Gould, W., 1993. Linear splines and piecewise linear functions. Stata Technical Bulletin 15, 13-17.
Graham, B., 2005. Intelligent Investor: The Classic Text on Value Investing. Harper Collins. (Original work
published in 1949).
Graham, J.R., Harvey, C.R., Puri, M., 2013. Managerial attitudes and corporate actions. Journal of Financial
Economics 109, 103-121.
Gregory, A., 1997. An examination of the long run performance of U.K. acquiring firms. Journal of Business
Finance & Accounting 24, 971-1002.
Gregory, A., 2005. The long run abnormal performance of U.K. acquirers and the free cash flow hypothesis.
Journal of Business Finance & Accounting 32, 777-814.
Gregory, A., O'Donohoe, S., 2014. Do cross border and domestic acquisitions differ? Evidence from the
acquisition of U.K. targets. International Review of Financial Analysis 31, 61-69.
Grinblatt, M., Keloharju, M., 2001. What makes investors trade? Journal of Finance 56, 589-616.
Grinstein, Y., Hribar, P., 2004. CEO compensation and incentives: Evidence from M&A bonuses. Journal of
Financial Economics 73, 119-143.
Guay, W.R., 1999. The sensitivity of CEO wealth to equity risk: An analysis of the magnitude and determinants.
Journal of Financial Economics 53, 43-71.
Guest, P.M., 2009. The impact of mergers and acquisitions on executive pay in the United Kingdom. Economica
76, 149-175.
Gugler, K., Mueller, D.C., Yurtoglu, B.B., Zulehner, C., 2003. The effects of mergers: an international comparison.
International Journal of Industrial Organization 21, 625-653.
Guo, J.M., Petmezas, D., 2012. What are the causes and effects of M&As? The U.K. evidence. Multinational
Finance Journal 16, 1-27.
Hagendorff, J., Collins, M., Keasey, K., 2008. Investor protection and the value effects of bank merger
announcements in Europe and the US. Journal of Banking & Finance 32, 1333-1348.
Hansen, R.G., 1987. A theory for the choice of exchange medium in mergers and acquisitions. Journal of Business
60, 75-95.
References
311
Harford, J., 1999. Corporate cash reserves and acquisitions. Journal of Finance 54, 1969-1997.
Harford, J., Humphery-Jenner, M., Powell, R., 2012. The sources of value destruction in acquisitions by
entrenched managers. Journal of Financial Economics 106, 247-261.
Harford, J., Klasa, S., Walcott, N., 2009. Do firms have leverage targets? Evidence from acquisitions. Journal of
Financial Economics 93, 1-14.
Harford, J., Li, K., 2007. Decoupling CEO wealth and firm performance: The case of acquiring CEOs. Journal of
Finance 62, 917-949.
Hartzell, J.C., Ofek, E., Yermack, D., 2004. What’s in it for me? CEOs whose firms are acquired. Review of
Financial Studies 17, 37-61.
Hayward, M.L., Hambrick, D.C., 1997. Explaining the premiums paid for large acquisitions: Evidence of CEO
hubris. Administrative Science Quarterly, 103-127.
Healy, P.M., Palepu, K.G., Ruback, R.S., 1992. Does corporate performance improve after mergers? Journal of
Financial Economics 31, 135-175.
Heath, C., Huddart, S., Lang, M., 1999. Psychological factors and stock option exercise. Quarterly Journal of
Economics, 601-627.
Heider, F., 1958. The psychology of interpersonal relations. New York: Wiley. p. 322.
Higson, C., Elliott, J., 1998. Post-takeover returns: The U.K. evidence. Journal of Empirical Finance 5, 27-46.
Hirshleifer, D., Low, A., Teoh, S.H., 2012. Are overconfident CEOs better innovators? Journal of Finance 67,
1457-1498.
Hoberg, G., Phillips, G., 2010. Product market synergies and competition in mergers and acquisitions: A text-
based analysis. Review of Financial Studies 23, 3773-3811.
Hodgkinson, L., Partington, G.H., 2008. The motivation for takeovers in the U.K. Journal of Business Finance &
Accounting 35, 102-126.
Houston, J.F., James, C.M., Ryngaert, M.D., 2001. Where do merger gains come from? Bank mergers from the
perspective of insiders and outsiders. Journal of Financial Economics 60, 285-331.
References
312
Huang, Y.-S., Walkling, R.A., 1987. Target abnormal returns associated with acquisition announcements:
Payment, acquisition form, and managerial resistance. Journal of Financial Economics 19, 329-349.
Huddart, S., Lang, M., Yetman, M.H., 2009. Volume and price patterns around a stock's 52-week highs and lows:
Theory and evidence. Management Science 55, 16-31.
Ismail, A., Krause, A., 2010. Determinants of the method of payment in mergers and acquisitions. Quarterly
Review of Economics and Finance 50, 471-484.
Jarrell, G.A., Poulsen, A.B., 1989. The returns to acquiring firms in tender offers: Evidence from three decades.
Financial Management, 12-19.
Jemison, D.B., Sitkin, S.B., 1986. Corporate acquisitions: A process perspective. Academy of Management Review
11, 145-163.
Jensen, M., 1986. Agency cost of free cash flow, corporate finance, and takeovers. Corporate Finance, and
Takeovers. American Economic Review 76.
Jensen, M.C., 1988. Takeovers: Their causes and consequences. Journal of Economic Perspectives 2, 21-48.
Jensen, M.C., 2005. Agency costs of overvalued equity. Financial Management 34, 5-19.
Jensen, M.C., Meckling, W.H., 1976. Theory of the firm: Managerial behavior, agency costs and ownership
structure. Journal of Financial Economics 3, 305-360.
Kahneman, D., 2003. Maps of bounded rationality: Psychology for behavioral economics. American Economic
Review 93, 1449-1475.
Kahneman, D., Tversky, A., 1979. Prospect theory: An analysis of decision under risk. Econometrica: Journal of
the Econometric Society 47, 263-292.
Kau, J.B., Linck, J.S., Rubin, P.H., 2008. Do managers listen to the market? Journal of Corporate Finance 14,
347-362.
Kim, J.-Y.J., Haleblian, J.J., Finkelstein, S., 2011. When firms are desperate to grow via acquisition: The effect
of growth patterns and acquisition experience on acquisition premiums. Administrative Science Quarterly 56, 26-
60.
King, D.R., Dalton, D.R., Daily, C.M., Covin, J.G., 2004. Meta‐analyses of post‐acquisition performance:
Indications of unidentified moderators. Strategic Management Journal 25, 187-200.
References
313
Kliger, D., Kudryavtsev, A., 2008. Reference point formation by market investors. Journal of Banking & Finance
32, 1782-1794.
Kohers, N., Kohers, G., Kohers, T., 2007. Glamour, value, and the form of takeover. Journal of Economics and
Business 59, 74-87.
Kuipers, D.R., Miller, D.P., Patel, A., 2009. The legal environment and corporate valuation: Evidence from cross-
border takeovers. International Review of Economics & Finance 18, 552-567.
Kumar, A., 2009. Who gambles in the stock market? Journal of Finance 64, 1889-1933.
Kumar, M., Dixit, J., Francis, B., 2015. The impact of prior stock market reactions on risk taking in acquisitions.
Strategic Management Journal 36, 2111-2121.
Lang, L.H.P., Stulz, R.M., Walkling, R.A., 1989. Managerial performance, Tobin's Q, and the gains from
successful tender offers. Journal of Financial Economics 24, 137-154.
Larcker, D.F., Rusticus, T.O., 2010. On the use of instrumental variables in accounting research. Journal of
Accounting and Economics 49, 186-205.
Lehn, K.M., Zhao, M., 2006. CEO turnover after acquisitions: Are bad bidders fired? Journal of Finance 61,
1759-1811.
Levis, M., 2011. The performance of private equity‐backed IPOs. Financial Management 40, 253-277.
Li, J., Yu, J., 2012. Investor attention, psychological anchors, and stock return predictability. Journal of Financial
Economics 104, 401-419.
Lin, H.-C., Chou, T.-K., Cheng, J.-C., 2011. Does market misvaluation drive post-acquisition underperformance
in stock deals? International Review of Economics & Finance 20, 690-706.
Lin, L., Lin, D., 2014. Agency costs of free cash flow and bidders’ long-run takeover performance. Universal
Journal of Accounting and Finance 2, 1-8.
Linn, S.C., Switzer, J.A., 2001. Are cash acquisitions associated with better postcombination operating
performance than stock acquisitions? Journal of Banking & Finance 25, 1113-1138.
Loderer, C., Martin, K., 1990. Corporate acquisitions by listed firms: The experience of a comprehensive sample.
Financial Management, 17-33.
References
314
Loderer, C., Martin, K., 1992. Post-acquisition performance of acquiring firms. Financial Management, 69-79.
Loughran, T., Ritter, J.R., 2002. Why don’t issuers get upset about leaving money on the table in IPOs? Review
of Financial Studies 15, 413-444.
Loughran, T., Vijh, A.M., 1997. Do long‐term shareholders benefit from corporate acquisitions? Journal of
Finance 52, 1765-1790.
Luo, Y., 2005. Do insiders learn from outsiders? Evidence from mergers and acquisitions. Journal of Finance 60,
1951-1982.
Lyon, J.D., Barber, B.M., Tsai, C.L., 1999. Improved methods for tests of long‐run abnormal stock returns.
Journal of Finance 54, 165-201.
Malmendier, U., Tate, G., 2005. CEO overconfidence and corporate investment. Journal of Finance 60, 2661-
2700.
Malmendier, U., Tate, G., 2008. Who makes acquisitions? CEO overconfidence and the market’s reaction. Journal
of Financial Economics 89, 20-43.
March, J.G., Shapira, Z., 1987. Managerial perspectives on risk and risk taking. Management Science 33, 1404-
1418.
Martin, K.J., 1996. The method of payment in corporate acquisitions, investment opportunities, and management
ownership. Journal of Finance 51, 1227-1246.
Martin, K.J., McConnell, J.J., 1991. Corporate performance, corporate takeovers, and management turnover.
Journal of Finance 46, 671-687.
Martynova, M., Oosting, S., Renneboog, L., 2006. The long-term operating performance of European mergers
and acquisitions.
Martynova, M., Renneboog, L., 2006. Mergers and acquisitions in Europe. SSRN Working Paper.
Martynova, M., Renneboog, L., 2008. A century of corporate takeovers: What have we learned and where do we
stand? Journal of Banking & Finance 32, 2148-2177.
Martynova, M., Renneboog, L., 2009. What determines the financing decision in corporate takeovers: Cost of
capital, agency problems, or the means of payment? Journal of Corporate Finance 15, 290-315.
References
315
Martynova, M., Renneboog, L., 2011. The performance of the European market for corporate control: Evidence
from the fifth takeover wave. European Financial Management 17, 208-259.
Massa, M., Zhang, L., 2009. Cosmetic mergers: The effect of style investing on the market for corporate control.
Journal of Financial Economics 93, 400-427.
Masulis, R.W., Simsir, S.A., 2015. Deal initiation in mergers and acquisitions. SSRN Working Paper.
Mehra, R., Prescott, E.C., 1985. The equity premium: A puzzle. Journal of Monetary Economics 15, 145-161.
Mitchell, M.L., Lehn, K., 1990. Do bad bidders become good targets? Journal of Political Economy 98, 372-398.
Moeller, S., 2015. Why taking risks in M&A should pay off.
Available at https://blogs.intralinks.com/2015/07/why-taking-risks-in-mergers-acquisitions-should-pay-off/.
Moeller, S.B., Schlingemann, F.P., 2005. Global diversification and bidder gains: A comparison between cross-
border and domestic acquisitions. Journal of Banking & Finance 29, 533-564.
Moeller, S.B., Schlingemann, F.P., Stulz, R.M., 2004. Firm size and the gains from acquisitions. Journal of
Financial Economics 73, 201-228.
Moeller, S.B., Schlingemann, F.P., Stulz, R.M., 2007. How do diversity of opinion and information asymmetry
affect acquirer returns? Review of Financial Studies 20, 2047-2078.
Morck, R., Shleifer, A., Vishny, R.W., 1988. Management ownership and market valuation: An empirical analysis.
Journal of Financial Economics 20, 293-315.
Morck, R., Shleifer, A., Vishny, R.W., 1990. Do managerial objectives drive bad acquisitions? Journal of Finance
45, 31-48.
Morrow, J., Sirmon, D.G., Hitt, M.A., Holcomb, T.R., 2007. Creating value in the face of declining performance:
Firm strategies and organizational recovery. Strategic Management Journal 28, 271-283.
Mueller, D.C., 1980. The determinants and effects of mergers: An international comparison. Oelgeschlager, Gunn
& Hain; Königstein/Ts.: Verlag A. Hain.
Myers, S.C., Majluf, N.S., 1984. Corporate financing and investment decisions when firms have information that
investors do not have. Journal of Financial Economics 13, 187-221.
References
316
Northcraft, G.B., Neale, M.A., 1987. Experts, amateurs, and real estate: An anchoring-and-adjustment perspective
on property pricing decisions. Organizational Behavior and Human Decision Processes 39, 84-97.
Odean, T., 1998. Are investors reluctant to realize their losses? Journal of Finance 53, 1775-1798.
Olsen, R.A., 1997. Prospect theory as an explanation of risky choice by professional investors: Some evidence.
Review of Financial Economics 6, 225-232.
Palepu, K.G., 1986. Predicting takeover targets: A methodological and empirical analysis. Journal of Accounting
and Economics 8, 3-35.
Pachare S., 2010, Market Power, Extrapolation and the Cross-Section of Stock Returns. Draft.
Pettengill, G.N., Jordan, B.D., 1990. The overreaction hypothesis, firm size, and stock market seasonality. Journal
of Portfolio Management 16, 60-64.
Powell, R.G., Stark, A.W., 2005. Does operating performance increase post-takeover for U.K. takeovers? A
comparison of performance measures and benchmarks. Journal of Corporate Finance 11, 293-317.
Raj, M., Forsyth, M., 2003. Hubris amongst U.K. bidders and losses to shareholders. International Journal of
Business 8.
Rau, P., Vermaelen, T., 1998. Glamour, value and the post-acquisition performance of acquiring firms. Journal
of Financial Economics 49, 223-253.
Rhodes-Kropf, M., Robinson, D.T., 2008. The market for mergers and the boundaries of the firm. Journal of
Finance 63, 1169-1211.
Rhodes-Kropf, M., Robinson, D.T., Viswanathan, S., 2005. Valuation waves and merger activity: The empirical
evidence. Journal of Financial Economics 77, 561-603.
Rhodes-Kropf, M., Viswanathan, S., 2004. Market valuation and merger waves. Journal of Finance 59, 2685-
2718.
Roll, R., 1986. The hubris hypothesis of corporate takeovers. Journal of Business 59, 197-216.
Rossi, S., Volpin, P.F., 2004. Cross-country determinants of mergers and acquisitions. Journal of Financial
Economics 74, 277-304.
References
317
Ruback, R.S., 1988. Do target shareholders lose in unsuccessful control contests? Corporate takeovers: Causes
and consequences: University of Chicago Press, pp. 137-156.
Sacheti, A., Gregory-Smith, I., Paton, D., 2016. Managerial decision making under uncertainty: The case of
twenty 20 cricket. Journal of Sports Economics 17, 44-63.
Sanders, W.G., Hambrick, D.C., 2007. Swinging for the fences: The effects of CEO stock options on company
risk taking and performance. Academy of Management Journal 50, 1055-1078.
Savor, P.G., Lu, Q., 2009. Do stock mergers create value for acquirers? Journal of Finance 64, 1061-1097.
Schneider, C., Spalt, O.G., 2015. Acquisitions as lotteries? The selection of target-firm risk and its impact on
merger outcomes (January 9, 2015). Critical Finance Review (Forthcoming).
Schwert, G.W., 1996. Markup pricing in mergers and acquisitions. Journal of Financial Economics 41, 153-192.
Schwert, G.W., 2000. Hostility in takeovers: In the eyes of the beholder? Journal of Finance 55, 2599-2640.
Servaes, H., 1991. Tobin's Q and the gains from takeovers. Journal of Finance 46, 409-419.
Seth, A., Song, K.P., Pettit, R., 2000. Synergy, managerialism or hubris? An empirical examination of motives
for foreign acquisitions of US firms. Journal of International Business Studies 31, 387-405.
Seth, A., Song, K.P., Pettit, R.R., 2002. Value creation and destruction in cross‐border acquisitions: an empirical
analysis of foreign acquisitions of US firms. Strategic Management Journal 23, 921-940.
Shefrin, H., Statman, M., 1985. The disposition to sell winners too early and ride losers too long: Theory and
evidence. Journal of Finance 40, 777-790.
Shleifer, A., Vishny, R.W., 1988. Value maximization and the acquisition process. Journal of Economic
Perspectives 2, 7-20.
Shleifer, A., Vishny, R.W., 2003. Stock market driven acquisitions. Journal of Financial Economics 70, 295-311.
Smith, A., 1759. The Theory of Moral Sentiment (London: A. Millar).
Soegiharto, S., 2009. What drives the payment of higher merger premiums? Gadjah Mada International Journal
of Business 11, 191-228.
References
318
Stock, J.H., Wright, J.H., Yogo, M., 2002. A survey of weak instruments and weak identification in generalized
method of moments. Journal of Business & Economic Statistics 20, 518-529.
Stulz, R., 1988. Managerial control of voting rights: Financing policies and the market for corporate control.
Journal of Financial Economics 20, 25-54.
Sudarsanam, S., Mahate, A.A., 2003. Glamour acquirers, method of payment and post‐acquisition performance:
The U.K. evidence. Journal of Business Finance & Accounting 30, 299-342.
Switzer, J.A., 1996. Evidence on real gains in corporate acquisitions. Journal of Economics and Business 48, 443-
460.
Trautwein, F., 1990. Merger motives and merger prescriptions. Strategic Management Journal 11, 283-295.
Travlos, N.G., 1987. Corporate takeover bids, methods of payment, and bidding firms’ stock returns. Journal of
Finance 42, 943-963.
Tversky, A., Kahneman, D., 1974. Judgment under uncertainty: Heuristics and biases. Science 185, 1124-1131.
Tversky, A., Kahneman, D., 1992. Advances in prospect theory: Cumulative representation of uncertainty.
Journal of Risk and Uncertainty 5, 297-323.
Uysal, V.B., 2011. Deviation from the target capital structure and acquisition choices. Journal of Financial
Economics 102, 602-620.
Uysal, V.B., Kedia, S., Panchapagesan, V., 2008. Geography and acquirer returns. Journal of Financial
Intermediation 17, 256-275.
Vermaelen, T., Xu, M., 2014. Acquisition finance and market timing. Journal of Corporate Finance 25, 73-91.
Vijh, A.M., Yang, K., 2013. Are small firms less vulnerable to overpriced stock offers? Journal of Financial
Economics 110, 61-86.
Wansley, J.W., Lane, W.R., Yang, H.C., 1987. Gains to bidder firms in cash and securities transactions. Financial
Review 22, 403-414.
Weitzel, U., Kling, G., 2014. Sold below value? Why some takeover offers have negative premiums. SSRN
Working Paper.
References
319
Wulf, J., 2004. Do CEOs in mergers trade power for premium? Evidence from “mergers of equals”. Journal of
Law, Economics, and Organization 20, 60-101.
Zhang, L., 2009. Why do firms pay cash in acquisitions? Evidence from a demand perspective. SSRN Working
Paper.
Zhang, P.M., 2001. What really determines the payment methods in M&A deals. Manchester School of
Management, Working Paper.